SUPERCRITICAL WATER OXIDATION MODEL DEVELOPMENT
FOR SELECTED EPA PRIORITY POLLUTANTS
.by
Earnest F. Gloyna, Principal Investigator,
and
Lbdong U, Task Manager
Separations Research Program
Center for Energy Studies
and
Environmental and Water Resources Engineering
Department of CM Engineering
The University of Texas at Austin
Austin TX 78712
EPA ASSISTANCE ID NO. CR-8t6760-02-0
Technical Project Monitor
Ronald J. Turner
Water and Hazardous Waste Treatment Research Division
Risk Reduction Engineering Laboratory
Cincinnati, OH 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI OH 45268
-------
NOTICE
Although the information in this document has been funded wholly by the United States!Environmental
Protection Agency under Assistance Agreement CR-816760-02 to the University of Texas at Austin, it
may not necessarily reflect the views of the agency and no official endorsement should be inferred.
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial products and practices
frequently carry with them the increased generation of materials that, if improperly 'dealt with, can
threaten both public health and the environment. The U.S. Environmental Protection Agency (EPA) is
charged by Congress with protecting the Nation's land, air, and water resources. Under a mandate of
national environmental laws, the Agency strives to formulate and implement actions leading to a
compatible balance between human activities and the ability of natural systems to support and nurture
life. These laws direct the EPA to perform research to define our environmental problems, measure the
impacts, and search for solutions. :
The Risk Reduction Engineering Laboratory is responsible for planning, Implementing, arid
managing research, development, and demonstration programs to provide an authoritative, defensible
engineering basis in support of the policies, program, and regulations of the EPA with respect to
drinking water, wastewater, pesticides, toxic substances, solid and hazardous wastes, and Superfund-
related activities. This publication is one of the products of that research and provides a vital
communication link between the researcher and the user community.
'• •••.-;•-• - - - - ' \ . I-'-, i, ,,<•:; •.", •
This effort presents the results of supercritical water oxidation (SCWO) of five compounds:,
acetic acid, 2,4-dichlorophenol, pentachlorophenol, pyridine, and 2,4-dichtorophenbxyacetic add'
(methyl ester). Kinetic models were developed for acetic acid, 2,4-dichlorophenol^ and pyridine.
Critical engineering issues were evaluated and the results were used to develop SCWO process design
and operating strategies, assisting in future commercialization of SCWO.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
ABSTRACT
Supercritical water oxidation (SCWO) kinetic models and reaction pathways for selected EPA
priority pollutants were developed. Critical engineering issues were evaluated, and tlie results were
used to develop SCWO process design and operating strategies.
The project was executed in three phases. The first phase (year one) involved batch studies.
SCWO destruction of five model compounds (acetic acid, 2,4-dichlorophenol, pentachlorophenol,
pyridfne. and 2,4-dichlorophenoxyacetic acid methyl ester) was evaluated. The second phase (year
two) consisted of detailed, continuous-flow tests (kinetic and mechanistic studies of 2,4^dichlorophenol
and pyridine). The third phase (years one and two) dealt with the evaluation of critical engineering
issues (corrosion and chromium speciatfon).
The test compounds were effectively destroyed. The five model compounds exhibited a wide
range of reactivity in SCWO environments, indicating the effect of chemical and structural features of
each compound on the overall reaction rate. Kinetic models were developed for 2,4-dichlorophenol,
pyridine, and acetic acid.
Mechanistic studies involving 2,4-dichlorophenol and pyridine provided an insight into the
possible reaction pathways and by-product transformation. The breakdown of complex organic
molecules under SCWO conditions produced a large number of unstable compounds and a small
numbsrof relatively stable, lower-molecular-weight, intermediate compounds. By adjusting the reaction
conditions, the type and amount of intermediates produced can be controlled, hence a^teving more
efficient reactor design and higher destruction efficiency. i >*•; ^fV:
For a given temperature, the highest corrosion rate occurred at the lowest pffin thSesf
conditions, which ranged from PH , 2.1 to PH = 8.6. For a given PH, higher corrosion rates were
observed at 300'C and 500'C, as compared to 400'C. The formation of chromium species can be
controlled by adjusting the pH. Hexavalent chromium can be precipitated effectively because of the
limited solubility of chromate salts in supercritical water.
This report is submitted in fulfillment of EPA Contract Number CR-816760^02-0 by The
University of Texas at Austin under the matching sponsorship of the U.S. Environmental Protection
Agency (EPA) and the Separations Research Program (SRP), The University of Texas at Austin. The
SRP is a consortium of about 30 industries. This report covers a period from October 1, 1990 to
December 31,1992. :
IV
-------
TABLE OF CONTENTS
Notice . g
Foreword g,-
Abstract -N
List of Figures . . vii
List of Tables « . . vii
Abbreviations and Symbols -K
Acknowledgments . . xi
Executive Summary : 1
1. Introduction 5
1.1 Supercritical water oxidation 5
1.2 Objective ...... 6
1.3 Scope 7
1.4 Rationale . . . ! ! ! | 7
2. Conclusions g
2.1 Batch and continuous-flow studies 9
22. Corrosion study ' '\ \ $
2.3 Chromium speclatton and separation study . . . . . ". . '.'•'. \. to
3. ^ Recommendations 1 . . . i . . 12
• ..j*C': ' . i-1-'-
3.1 Batch and continuous-flow studies . . . 12
32 Corrosion study 12
3.3 Chromium spedatfon and separation study . . . . . . . ].! '. .13
3.4 Erosion evaluation ! 13
4. Materials and Methods i . . 14
4.1 Test plans i . . 14
42 Tasks and milestones ; . ! 15
4.3 Quality assurance project plan (QAPP) ; ! ! 15
4.4 Materials ; ... 17
5. Experimental Procedures ; 19
5.1 Test procedures 19
5.1.1 Batch system #1 ; . . 19
5.1.2 Batch system #2 i . ! 22
5.1.3 Continuous-flow system #1 ! . 25
5.1.4 Continuous-flow system #2 i . 27
52 Feed preparation . : . . 28
5.3 Sample collection and analyses . 29
5.3.1 Sample collection ! 29
-------
5.3.1 Sample collection . . 29
5.3.2 Gas chromatography (GC) . . 29
5.3.3 GC-mass spectroscopy (GC-MS) . . 30
5.3.4 Ion chromatography (1C) 30
5-3.5 pH 30
5.3.6 Chemical oxygen demand 31
5.3.7 Ammonia . 31
5.3.8 Total suspended solids 31
5.3.9 Liquid chromatography .31
5.3.10 Chromium species : . ! 31
6. Results and Discussion . 33
6.1 Data quality 33
62 Batch test ; [ 33
6.2.1 Acetic acid . ... 33
6.2.2 Pentachlorophenol (PCP) '• 34
6.2.3 2,4-DichlorophenoI (2,4-DCP) . . 37
6.2.4 Pyridine . . 39
6.2.5 2,4-Dichlorophenoxyacetic acid methyl ester .39
6.3 Continuous-flow test 44
6.3.1 2,4-Dtohlorophenol . 46
6.3.2 Pyridine '..'.! 50
6.3.3 Summary of kinetic models . . . ! 60
6.4 Corrosion tests I . ] $-\
6.5 Chromium speciation and separation , .63
6.6 Erosion evaluation . 64
6.7 Process development ! ! 65
7. Information Transfer 67
!
References . . . • . . . , . . ' .68
Appendices -
A. List of publications and presentations . ... . . . . .", . .72
B. Photographs of experimental setups . . . . . . . . . . | 73^
VI
-------
UST OF FIGURES
4.1 Project Tasks and Milestones 16
5.1 Batch Reactor System #1 .20
5.2 Batch Reactor Assembly #1 . . 21
I
5.3 Batch Reactor System #2 23
5.4 Batch Reactor Assembly #2 ; . . 24
5.5 Laboratory-Scale, Continuous-Flow SCWO Reactor System #1 ...;.. 26
6,1 Comparison of 2,4-DCP Oxidation Conversion: Predicted vs. Observed .' . .49
6.2 Percent of Pyridine Remaining During SCWO , . .54
6.3 Simplified Reaction Pathways for Pyridine Oxidation in SCW—Ring Opening 'Step
Involving 2-Hydroxypyridine Intermediate and Carbon-Nitrogen Bonds . . . .57
6.4 Simplified Reaction Pathways for Pyridine Oxidation in SCW—Ring Opening Step
Involving 2-Hydroxypyridine Intermediate and Carbon-Carbon Bonds . . . .58
6.5 Simplified Reaction Pathways for Pyridine Oxidation in SCW—Ring Opening Step
Involving 4- and_3-Hydroxypyridine intermediates . . 59
- UST OF TABLES
6.1 Summary of Data Quality „ . . . , . . . .. ... . . .34
-~* - .* -j'-v^i. *' .'^'v **• • " • » • '
6.2 Sample Analytical Report for 1C Measurements .. . . . . . . .35
6.3 Sample Analytical Report for GC Measurements ......... 36
6.4 Summary of Chromatographic Analyses, Calibration Methods,
and Detection Limits 38
6.5 Summary of Batch Test Conditions 39
6.6 Acetic Acid Batch Test Results : . . 40
6.7 Pentachlorophenol Batch Test Results . . . . . . . . . I . .41
6.8 2,4-Dichlorophenol Batch Test Results . 42
6.9 Pyridine Batch Test Results , i . . 43
6.10 2,4-Dichlorophenoxyacette Acid Methyl Ester Batch Test Results . . . . .45
6.11 Continuous-Flow Test Conditions for 2,4-Dichlorophenol .....;.. 46
6.12 Continuous-Flow Test Data for 2,4-Dichlorophenol . 47
6.13 2,4-DCP Oxidation Transition Products . 51
vu
-------
6.14
6.15
6.16
6.17
6.18
6.19
6.20
6.21
Continuous-Flow Test Conditions for Pyridine 51
Continuous-Flow Test Data for Pyridine . . 52
Pyridine Oxidation Transition Products Identified by 1C . 55
Pyridine Oxidation Transition Products Identified by GC 56
Summary of SCWO Kinetic Models Derived from This Work and the Literature . . 60
Uniform Corrosion Rates for Three Alloys in Subcritical and Supercritical Water. . 62
Chromium Concentrations Derived from Sub-CWO and SCWO
of Anaerobically Digested Municipal Sludge 64
Chromium Concentrations Derived from Sub-CWO and SCWO
of Industrial Sludge \
Vlll
-------
Abbreviations
COD
QC
GC-MS
gph
HPLC
1C
D
MMT
mm/y -
mpy
00
PCP
QA
QAPP . -
QC
RF
RSD
SRP
SCW
SCWO
SS
TETF
2,4-D
2,4-DCP
TSS
ABBREVIATIONS AND SYMBOLS
chemical oxygen demand
gas chromatography
gas chromatography-mass spectroscopy
gallons per hour
high performance liquid chromatograph
ion chromatography
inner diameter
million metric tons
millimeters per year
mils per year
outer diameter
pentachlorophenol
quality assurance
quality assurance project plan
quality control
response factor
relative standard deviation
Separations Research Program (UT-Austin)
supercritical water
supercritical water oxidation
Stainless Steel
totally enclosed treatment facility
2,4-dfehIorophenoxyacetic acid
2,4-dichIorophenol
total suspended solids
IX
-------
Symbols
a, b,c
A
C
Ea
k
Reaction orders
Pre-exponential factor
Concentration
Activation energy
Reactor rate constant
r
R
T
X
Solubility product constant
Reaction rate
The gas law constant
Temperature
Conversion, X = 1-Cf£o
-------
ACKNOWLEDGMENTS
Funds were provided under a matching contract between the U.S. Environmental Protection
Agency (EPA) Risk Reduction Engineering Laboratory (RREL), Cincinnati, Ohio, and the Separations
Research Program (SRP), The University of Texas at Austin. The research was conducted under U.S.
EPA Grant No. CR816760-02-0. Additional funds were provided by the Bettie M. Smith Chair in
Environmental Health Engineering, Civil Engineering Department, The University of Texas at Austin.
Acknowledgment is extended to E. Timothy Oppelt, Director, and Ron Turner, Technical Project
Monitor, U.S. EPA RREL, and Dr. James Fair, Head, SRP.
Special recognition is extended to Dr. Dong Soo Lee and graduate students who performed
much of the research: Neil Grain, Anusuya Kanthasamy, Chris Matthews, Sam Rollans, and Dave
Sheets.
Selected projects that provided supporting data and equipment included the Energy Research in
Applications Program (State of Texas) and the Gulf Coast Hazardous Substances Center (EPA/TX),
Beaumont, Texas.
XI
-------
EXECUTIVE SUMMARY
The research described herein was part of a two-year, matching assistance agreement for the
period of October 1 , 1 990 to December 31 , 1 992, inclusive.
The goal of this project was to assess the performance of supercritical water oxidation (SCWO)
in treating selected EPA priority pollutants and to enhance the development of SCWO waste
treatment processes.
The project was executed in three phases. The first phase (year one) involved batch studies:
SCWO destruction of five model compounds (acetic acid, 2,4-dichlorophenol, pentachlorophenol,
pyridine, and 2,4-dichlorophenoxyacetic acid methyl ester). The second phase (year two) consisted of
detailed, continuous-flow tests involving both kinetic and mechanistic studies of 2,4-dichlorophenol
(2,4-DCP) and pyridine. The third phase (years one and two) dealt with the evaluation of critfcati
engineering issues such as corrosion and chromium speciation.
Batch Study: Batch experiments were carried out at three temperatures (400°C, 450°Cf 500?C). a
constant water density (0.3 g/mL), and reaction times varying from 2 min to 20 mim tMU-shaoed '
-" ." ^r-' /-. _ | '•• "*•••_ • .' -f: - -•** ;^&;~ ^\'*3*&&&f*i -^ -'•£?, .-
reactors (volume = 20 mL) were made of Stainless Steel (SS) Sistubing wi»kb.8i?cm j"
(LD.) and a 1.27 cm outer diameter (O.D.). Heat was provided by a fluJdized I san
sand bath, each reactor was mechanically vibrated to enhance mixing and heat transfer. Feed
concentrations ranged from 40 mg/L to 3000 mg/L, and oxygen was used as the oxidant. The model
compounds were analyzed by chromatographic techniques.
Destruction efficiencies of >99.99% were observed for pentachlorophenol at a temperature of
500"C and a reaction time of 2 min. Destruction efficiencies of >99% were observed for 2,4-DCP ami
2,4-D methyl ester at a temperature of 500°C and a reaction time of 10 min. Acetic acid and pyridine, as
compared to the chlorinated aromatics, were relatively refractory, but destruction efficiencies of >99%
were observed at a temperature of 500°C and a reaction time of 20 min. Qualitatively, as determined by
GC analyses, SCWO of 2,4-DCP and 2,4-D methyl ester produced noticeable amounts of intermediate
compounds (about 20) at either lower temperatures (<450°C) or shorter reaction times (<5 min), and
the number of these intermediates reduced to about three when the temperature and reaction times
were >450°C and >5 min. SCWO of acetic acid, 2,4-DCP, and pyridine followed pseudo-first-order
-------
reaction kinetics. Pyridine and 2,4-DCP were recommended for more detailed kinetic and mechanistic
studies involving continuous-flow SCWO reactor systems.
Continuous-Flow Study: The continuous-flow experiments were conducted using a plug-flow
reactor system at temperatures varying from 400°C to 520°C, residence times ranging from 2 sec to 11
sec, >200% excess oxygen, and a pressure of 27.6 MPa. Reynolds number ranged from 7400 to
8200. Feed concentrations varied from 300 mg/L to 800 mg/L for 2,4-DCP and from 1000 mg/L to
3000 mg/L for pyridine. The reactor was made of SS 316 tubing (0.635 cm O.D. and 0.165 cm wall
thickness). The required reactor residence times were obtained by adjusting the tubing length while
maintaining the feed flow rate at 35 g/min.
About 10% of the 2,4-DCP was hydrolyzed by supercritical water at temperatures above
450°C. The rate of hydrolysis was first-order with respect to the concentration of 2,4-DCP. The
activation energy and pre-exponential factor were 209 kJ/mole and 1012-2 sec'1, respectively. The
overall oxidation and hydrolysis reaction rate (r) for 2,4-DCP was found to be r = A exp(-Ea/RT) [2,4-
DCP][O2]°:35, where the activation energy (Ea) and pre-exponential factor (A) were 88,9 kJ/mole and
105-5sec-1(mole/L)-°-35, respectively. Nine intermediate compounds were identified for 2,4-DCP
oxidation. These compounds included 2-chlorophenol, 4-chlorophenol. 2,6-dichIorophenol, phenol,
chloride, acetic acid, formic acid, carbon dioxide, and carbon monoxide. Based on these compounds,
a simplified reaction pathway for the SCWO of 2,4-DCP was developed. Less than 5% of the pyridino
was hydrolyzed by supercritical water at the highest tested temperature, 522°C. Therefore, .the SCWO
rate was approximated by the overalf oxidation and hydrolysis reaction rates.. The SC\^6for pyridino
was found to be r = A exp(-Ea/RT) (pyridfneMO^2, where the activation energy -1(JEa) and ^re-
exponential factor (A) were 210 kJ/mote and 1013-1 sec^moIe/L)-0-2, respectively. Seventeen
intermediate compounds were found for pyridine oxidation. These compounds included carboxylic
acids, dicarboxylic acids, amines, ammonia, carbon dioxide, and carbon monoxide. Based on the
presence of these compounds, a simplified reaction pathway for the SCWO of pyridine was
developed.
Corrosion Study: Three nickel alloys (SS 316, Hastelloy C-276, and Monel 400) were evaluated.
The experiments were conducted using a batch reactor setup at three temperatures (300°C, 400°C,
and 500°C), three pH conditions (2.1, 5.8, and 8.6), varying water densities (0.09 g/ct to 0.3 g/cc),
fixed oxygen loading (2.1 MPa), constant chloride concentration (420 mg/L), and a fixed exposure*
time (100 hrs). !
Both localized (pitting and crevice) and uniform corrosion were observed in all three alloys, and
additionally, selective leaching of the Monel 400 alloy was apparent at 400°C to 500°C. For both
-------
300°C or 500°C. Generally, the lowest pH condition, 2.1, created the most severe corrosion. For SS
316, the worst corrosion, 1.89 mpy, corresponded to a temperature of 300°C and pH of 2.1. For
Hastelloy C-276, the worst corrosion, 1.33 mpy, corresponded to a temperature of 500°C and pH of
5.8.
Chromium Speciation/Separation: The chromium speciation and separation study was
conducted using an existing vertical, concentric-tube reactor. The reactor was made of SS 316, which
contained 16 wt% chromium. Tests were conducted at varying temperatures (300°C to 450°C), feed
flow rates (45 g/min to 120 g/min), and a fixed pressure (25 MPa). Chromium concentrations of the
influent and effluent were monitored. Both municipal and industrial sludges were treated.
The chromium speciation, trivalent (Cr*3) and hexavalent (Cr+6), showed a direct correlation
with the effluent pH. When the effluent pH was less than 7, Cr*3 was the only detectable chromium
species in the treated effluent. Conversely, when the effluent pH was greater than 7, both Cr+3 and
Cr*6 corrosion products were found in the effluent. The level of Cr*6 in the treated effluent decreased
more than 10 times, 0.046 mg/L to 0.004 mg/L, when the process temperature changed from 3QO°C
to 400°C. At 400°C, the Cr+6 concentration at the reactor bottom (0.288 mg/L) was nfiuch higher, as
compared to the effluent Cr*6 concentration (0.004 mg/L), whereas at 300°C, the Cr+f concentration
at the reactor bottom (0.035 mg/L) was comparable with the effluent Cr*6 concentration (0.046 mg/L).
The Cr+3 concentration in the treated effluents decreased only 50%, 0.39 mg/L to OJ16 mg/L, whon
the process temperatures, changed from 300°C to 400°C. Jne precipitation of Qr*!8 was due to a
-'• .'-"••' -'.', • ' •"- .'••• ,•- .' '-• • -. "•• • • %l'-' L •; .
substantial decrease in the solubility of chromic and chromate salts. The Cr™ removal by precipitation
,,.,. ,- .-;•'.4grt.- . .;~.>£. • ±^7_ -v-V .-••-• .*--* /. ..-**-.;-• --**•*;/• •••
was affected by temperature, presence of specific co-foh& and co-ion concentration. Soluble Cr*3
was retained in the mass which settled in the reactor bottom. Co-precipitation with insoluble and
associated soluble salts appeared to be the mechanism by which the Cr*3 was removed.
A summary of the conclusions and recommendations is presented as follows:
• Refractory and chlorinated organic compounds, such as acetic acid, 2,4-dichlorophenol,
pentachlorophehol, pyridine, and 2,4-dichlorophenoxyacetic acid methyl ester, were
effectively destroyed by the SCWO process.
• These five model compounds exhibited a wide range of reactivity in SCWO environments,
indicating the effect of chemical and structural features of each compound on the overall
reaction rate. ,
• Kinetic models were developed for 2,4-dichlorophenol, pyridine, and acetic acid.
3
-------
• Mechanistic studies involving 2,4-dichlorophenol and pyridine provided an Insight into the
possible reaction pathways and species transformation.
• The SCWO of complex organic molecules produced a large number of unstable transition
compounds and a small number of refractory transition compounds.
i
• By adjusting the reaction conditions, the type and amount of transition compounds produced
were controlled, resulting in more efficient reactor design and higher destruction efficiency.
• For a given temperature, the highest corrosion rate occurred at the lowest pH (the pH range
was 2.1 to 8.6). For a given pH, higher corrosion rates occurred at 300°C and 500°C (the
temperature range was 300°C, 400°C, and 500°C).
• The formation of chromium species (trivaient and hexavalent) was affected by the pH of the
reaction media. ,
• Chromium species, hexavalent in particular, were precipitated effectively because of the
limited solubility of chromate salts in supercritical water.
-------
1. INTRODUCTION
Supercritical water oxidation (SCWO) of organic wastewaters and sludges has great potential
as a totally enclosed and acceptable treatment system capable of achieving any level of organic waste
destruction, thereby reducing the cost of waste management. The overall task of this project was to
demonstrate the applicability of the SCWO process, to develop kinetic data for selected EPA priority
pollutants, and to evaluate SCWO process performance. While the initial proposal was directed
towards deep-well subcritical and supercritical water oxidation reactor systems, this research was
modified to involve broader applications. Specific EPA listed priority pollutants were included in the
test program, and selected engineering tasks were specified for in-depth analyses. i
The following section presents a general discussion of the SCWO literature, research
objectives, scope, and rationale.
1.1 Supercritical water oxidation
The SCWO process utilizes the unique properties of water exhibited at conditions above its
vapor-liquid critical point (374°C and 22.1 MPa). Supercritical water (SOW) is an excellent solvent for
organic compounds (Connolly, 1966; Thomason and Model!, 1984) and becomes completely misclble
with oxygen (Japas and Franck, 1985). In the presence of supercritical water, hydrocarbons react
rapidly with oxygen to form carbon dioxide and water (Wightman, 1981; ^odefl et al., 1982; Wiirnanns
et al., 1989). Laboratory-scale studies on SCWO of wastewaters (Smith and Raptis, 1986; Takahashi
et al., 1989; Gloyna et al., 1990; Bramlette et al.. 1990) and organic sludges (Shanableh, 1990;
Tongdhamachart, 1^90; Modell, 1990) have demonstrated rapid and effective treatment.
The SCWO environment exhibits several unique characteristics. First, the SCWO rates are
likely to be controlled by reaction kinetics instead of the mass transfer rate (Wightman^ 1981; Helling,
1986; Helling and Tester, 1987; Helling and Tester, 1988r Yang and Eckert, 1988; Rofer and Strelt,
1989; Lee, 1990; Webley et al., 1991; Lee et al., 1991; Thornton and Savage, 1990; 1992). Second,
the solubility of inorganic safts in supercritical water is greatly reduced (Martynova, 1976; Bischoff et all.,
1986;; Dell'Orco and Gloyna, 1991; Armellini and Tester, 1991). Third, solid residuals resulting from
SCWO processes can meet EPA teachability requirements. These unique SCWO features make it
possible to produce acceptable gaseous, liquid, and solid effluents. Therefore, when influent wastes
are too dilute to incinerate economically, too toxic to treat biologically, or just too environmentally
sensitive, SCWO becomes an attractive treatment/destruction option (Freeman, 1985).
-------
The concept of applying SCWO to wastewater treatment was established more than a decade
ago (Thiel et at., 1979; Modell et al., 1982). A generic SCWO process involves an aqueous feed
stream containing liquid organic substances or a pumpable organic sludge, an oxidant source, and
possibly supplemental fuel or other additives. Following the SCWO reaction,: solid/liquid and
vapor/liquid separations are required. The oxidant may be air, oxygen, hydrogen peroxide, or perhaps
a compound containing oxygen. In cases where the heating value of the waste is insufficient,
supplemental fuel may be substituted for an external heat source. By using properly designed heat
exchangers, and a waste containing 2% to 5% organics, the heat developed by the oxidation can
maintain the required reaction temperature. Supplemental fuel may be relatively inexpensive orgsinic
wastes such as biological sludges, waste oils, and other organic compounds. The! concentration of
these wastes must be capable of sustaining SCW conditions. Conversely, if the wastes are too
concentrated, the influent must be diluted. In this case, the heat load to the process may exceed
design expectations of the materials of construction. ,
Various technical aspects of SCWO research and development activities were reviewed (Shaw
et al., 1991; Tester et al., 1992; Gloyna and Li, 1993). In addition, a comprehensive literature search
was conducted during the course of this project. The objective was to consolidate pertinent
information on wet oxidation. This comprehensive bibliography contains about 900 references
relating to all practical aspects of wet oxidation. The subject keywords include: reaction pathway,
kinetics, catalytic, solids separation, inorganic salt solubility, corrosion, erosion, reactor modeling,
wastewater, and sludge. .
Although some SCWO data were generated during the past ten years, most of these data
. ' " - '•-" , " • -" ' Y , " I*.'-** „ v
were produced to demonstrate high destruction efficiencies. Detailed information about SCWO
kinetics and reaction pathways for specific organic compounds was needed to assist reactor design.
Also, technological issues, such as erosion, corrosion, heavy metal speciation, heat transfer, solids
separation, and oxidant mixing, needed to be addressed.
!
1.2 Objective
The objective of this research was to expand the SCWO chemistry and engineering data base.
Specific objectives were: (1) conduct an extensive literature survey; (2) develop a quality assurance
project plan; (3) provide the destruction data for five organic compounds (four to be selected from the
U.S. EPA priority pollutant list); (4) conduct kinetic tests for two compounds based bn the screening
test results; (5) identify and test selected engineering problems associated with the SCWO process;
and (6) institute an effective technology transfer program. :
-------
1.3 Scope
All tests were conducted using either laboratory-scale, batch, or continuous-flow SCWO
reactor systems. Oxygen was used as the oxidant. Acetic acid, pentachlorophenol (POP), 2,4-
dichlorophenol (2,4-DCP), pyridine, and 2,4-dichlorophenoxyacetic acid (2,4-D) methyl ester were
I
used as the model compounds. The engineering problems under investigation and evaluation were
chromium speciation/separation, corrosion, and erosion.
1.4 Rationale
i
Clearly, the present toxic organic waste problem and sludge volume reduction challenge are
overwhelming if left to conventional treatment processes. Based on 1984 estimates extrapolated to
1990, hazardous wastes produced by industry range from 280 million metric tons per year (MMT/yr) to
395 MMT/yr. Hazardous solids constitute 2.27 MMT, and hazardous liquid wastes amount to 394 x 106
cubic meters (rr»3/yr). The U.S. municipal sludges contribute nearly 20 x 1Q6 dry T/yr.: The amount of
solids generated by conventional wastewater treatment processes (primary sedimentation and
secondary biological) may be in excess of 0.179 kg/m3 (1,500 Ib/million gallons) treated. About 1.01 x
108 m3 (27 billion gallons) and 1.1 x 106 m3 (290 million gallons), respectively, of municipal and toxic
liquid wastes are generated daily. Also, industry treats about 6.0 x 108 m3/d (160 million gallons per
day) of wastewater using biological treatment processes. In 1987 eleven waste disposal firms handled
4.6 MMT (5.07 x 106 tons) of waste: incineration (10.4%); resource recovery (7,2%); deep-woll
injertion (6.0%); chemical-biotogteaf treatment (22.2%); and landfill (54.2%). ;
While waste minimization is an important goal, there remain industrial and munici
that must be treated, recycled, and reused. Only two options exist for the complete
toxic organic wastewaters and sludges. These processes are incineration and supercritical water
oxidation.
Incineration with complete removal of stack discharges and ash stabilization has the potential
of completely burning all organic material. However, the problems with incineration have been: (a) high
construction and operating costs; (b) high temperature requirements, complete mixing of waste and
air, and assured residence time in the fire box; (c) need for high concentrations of organic wastes; (d)
redundancy needs in treatment of stack gases and ash handling; and (e) public opposition toward
incineration. Yet, the need for incineration is anticipated to grow as the new sludge regulations are
implemented. i
An alternative destruction and volume reduction process is SCWO. This process offers
several attractive advantages. SCWO reactions become self-sustaining with wastes haying a chemical
oxidation demand (COD) of about 30,000 mg/L In contrast, incineration requires an organic content
-------
of about 300,000 mg/L of COD. The SCWO concept can provide a totally enclosed treatment facility
that is environmentally safe. The costs appear to be favorable. !
The task of pollution control and abatement has moved well beyond the focus on technology-
based control of conventional point-source pollutants. Today, the national goals for hazardous waste
management include destruction levels of 99.9999% and containment systems that are "Totally
Enclosed Treatment Facilities" (TETF). The SCWO process can accomplish these goals and offer
unique and innovative solutions to costly waste management problems.
-------
2. CONCLUSIONS
The results obtained from this study support the following conclusions:
',
2.1 Batch and continuous-flow studies
• Refractory and chlorinated organic compounds, such as 2,4-dichlorophenol, penta-
chlorophenol, pyridine, and 2,4-dichlorophenoxyacetic acid methyl ester, were effectively
destroyed by the SCWO batch process. Similarly, acetic acid and other refractory transition
products of the SCWO operations were destroyed in the batch studies.
• These five model compounds exhibited a wide range of reactivity in SCWO environments,
indicating the effect of chemical and structural features of each compound on the overall
reaction rate.
• Kinetic models were developed for 2,4-dichlorophenol, pyridine, and acetic acid.
• Mechanistic studies involving 2,4-dichiorophenol and pyridine provided an insight into the
possible reaction pathways and reaction species transformation.
• Under some SCWO conditions, the breakdown of complex organic molecules was
accomplished by successive reduction of the molecular structure. This rapid chain reaction
transformed the initial organic mass into a large number of unstable compounds, an
intermediate group of relatively stable, low-molecular-weight compounds, and finally carbon
dioxide.
i
2.2 Corrosion study
• Metal coupons in the lowest (300°C) and highest temperature (500°C) experiments,
regardless of pH, generally experienced significantly more corrosion than those exposed to
the mid-range temperature (400°C). For example, the SS 316 specimens exposed to a pH of
5.8 and temperatures of 300°C and 500°C exhibited average corrosion rates of 1.63 and 0.36
!
mpy, respectively. The SS specimens exposed to a pH of 5.8 and 400°C showed an average
corrosion rate of 0.03 mpy.
-------
• Metal coupons exposed to the lowest pH test solution, regardless of temperature, generally
showed more corrosion than those exposed to the mid-range and high pH test solutions. For
example, Hastelloy C-276 specimens exposed to a pH of 2.1 and 300°C exhibited a uniform
corrosion rate of 0.64 mpy. Hastelloy C-276 specimens exposed to 300°C and pH levels of
5.8 and 8.6 exhibited corrosion rates of 0.46 and 0.48 mpy, respectively.
• Localized corrosion (pitting and crevice) was apparent All three alloys showed signs of this
type of corrosion.
• For wastes with low chloride concentrations and a neutral pH, SS 316 may be used as an
SCWO reactor material. If higher concentrations of aggressive ions are anticipated, a nickel-
chrome altoy such as Hastelloy C-276 may be considered.
• Monel 400 was not acceptable as a possible SCWO reactor material. For example, the uniform
corrosion rates of Monel 400 ranged from 1 to 27 mpy. Selective leaching was apparent in the
Monel 400 specimen.
2.3 Chromium speciation and separation study
• The pH of the effluent was an important factor in determining the oxidation state of the
chromium corrosion products. At a pH of <7, trivalent chromium was the only chromium
corrosion species generated. If the pH was >7, both trivalent and hexavalent chromium
corrosion species were generated. '
• Chromium species, hexavalent in particular, were precipitated effectively because of the
limited solubility of chromate salts in supercritical water. '
• The removal of hexavalent chromium from these effluents was temperature dependent. At
300°C and at a feed concentration of 0.035 mg/L, chromate salts were not removed from the
effluents. However, at 400°C these salts were removed by association and precipitation to
concentrations of <0.004 mg/L.
• Soluble trivalent chromium was removed from sub-CWO and SCWO processes as an impurity
by co-precipitation with insoluble and associated soluble salts. Under ambient conditions,
soluble trivalent chromium that co-precipitated with associated soluble salts was soluble.
Conversely, soluble trivalent chromium that co-precipitated with insoluble salts remained in the
insoluble phase.
10
-------
The removal rate of soluble trivalent chromium from these effluents was dependent on the
temperature and anion concentration. At 400°C, as compared to 300°C, the removal rate was
_p
larger by a factor of approximately five. Low concentrations of 804 in the effluent at SCWO
conditions reduced the likelihood of co-precipitation of soluble trivalent chromium with
associated soluble salts.
A substantial amount of clay was present in the industrial sludge. Soluble trivalent chromium
was not adsorbed to the oxidized clay.
11
-------
3. RECOMMENDATIONS
The direction of future SCWO research and process development is outlined by the following
recommendations.
3.1 Batch and continuous-flow studies i
• Generation of transition products should be considered in the design of a SCWO treatment
facility. Low-molecular-weight carboxylic acids may be produced in SCWO organic
compounds. Similarly, ammonia may be produced in SCWO of pyridine or other nitrogen-
containing organic compounds. ;
• SCWO treatment of chlorinated phenols and other organic compounds may generate a
significant amount of chloride ions. This tow pH (3.2 to 3.8) caused extensive corrosion of the
reactor and effluent cooler. Design of a SCWO reactor for the treatment of chlorinated
hydrocarbons must include corrosion resistant materials or use caustics to neutralize the
reaction fluid.
3.2 Corrosion study
• Consideration should be given to designing the reactor system to operate at temperatures just
above the critical point of water (about 400°C). This research has shown that corrosion rate
increases up to the critical point of water and drops off just above the critical point. As
temperature is further increased above the critical point of water to about 500°C, the corrosion
rate starts increasing.
• Further studies should examine the corrosion effects on materials such as titanium and
zirconium alloys, which do not contain chromium.
• Pre-stressed and welded specimens, respectively, should be used to examine stress
corrosion cracking and metal sensitization under SCWO conditions. These specimens should
be examined using cross-sectional analyses. :
12
-------
3.3 Chromium speciation and separation study i
» This research utilized a small-scale continuous-flow reactor. The flow regimes for these
experiments were in the transition and low-turbulent ranges. Consequently, i insoluble and
associated soluble salts were permitted to settle and collect in the reactor bottoms.
Commercial-scale SCWO reactors will operate at much higher Reynolds numbers and use
conventional solid separations technologies. Studies are required to determine if soluble
chromium species are removed effectively from commercial-scale SCWO effluents.
» The solubility and Ksp values for soluble trivalent and hexavalent chromium salts at different
SCWO conditions are unknown. These values are needed to predict the association and
precipitation of trivalent and hexavalent salts.
« The removal of trivalent and hexavalent soluble chromium species from SCWO effluent is
strongly influenced by the particular co-ion and co-ion concentration. Some cff these co-ions
form insoluble or low-solubility chromium salts at SCW conditions. Tradeoffs jof adding such
ions to sludge influent streams for the purpose of removing soluble chromium 'species should
be evaluated.
• Wastewater sludges containing high chromium concentrations should be studied.
3.4 Erosion evaluation
• Erosion of pressure letdown devices for SCWO applications can be reduced or minimized if
proper valve designs and materials of construction are selected.
• Pilot-scale pressure letdown testing is recommended. Impact of both solid and gaseous
components on erosion should be studied.
13
-------
4. MATERIALS AND METHODS
This section provides general research methodologies (such as test plans, tasks, and
milestones) and materials used in this study.
4.1 Test plans ;
The research plan consisted of four phases: (1) screening tests; (2) kinetic study; (3)
engineering evaluation; and (4) technology development and transfer. Details of the four phases are
described below.
Phase 1—Screening Tests
The purpose of the screening tests was to select two compounds for more rigorous kinetic
study. Information on the destruction efficiency and transition product formation under typical SCWO
conditions was needed. The proposed compounds for the screening tests included acetic acid,
pentachlorophenol, 2,4-dichlorophenol, pyridine, and 2,4-dichlorophen-oxyacetic acid methyl ester.
These compounds represented refractory compounds with different molecular structures (straight
chain and aromatic). All compounds, except for acetic acid, were either EPA priority pollutants or very
toxic compounds commonly found in wastewaters. Acetic acid was selected because jt was the most
stable transition product In thermal wastewater treatment processes and its oxidation was often the
rate-controlling step of the SCWO process.
The screening tests were conducted using a batch reactor system. Reaction temperature arid
presssure ranged from 400°C to 500°C and from 24 MPa to 34 MPa. Sample analysis were made mainly
using chromatographic techniques. Total organic carbon was also measured to investigate the
completeness of organic conversion. !
Phase 2—Rigorous Kinetic Study
In Phase 2, a detailed kinetic study was conducted for two compounds (2,4-DCP and pyridine)
using a continuous-flow reactor system. The purpose of continuous-flow tests was to develop global
kinetic models, identify transition products, and establish critical reaction pathways. Activation energy,
frequency factor, and reaction order with respect to reactant concentration were determined. In
addition, information on the interaction among the starting compound, water, and oxygen was
14
-------
obtained through transition product analyses. These kinetic and mechanistic data provided valuable
insight into SCWO process design, operation, and monitoring. i
A laboratory-scale, continuous-flow reactor system was constructed and used for the
experiments. The compounds were tested at temperatures ranging from 350°C to 53p°C. Pressure
was fixed at 27.6 MPa. The oxygen concentration was maintained within a range of 0% to 400% of the
theoretical oxygen demand. The conversions of the starting compounds were observed at four or five
retention times at each reaction temperature. In addition to the liquid sample analyses as described in
Phase 1, gaseous effluent samples were analyzed. All the conversion data were analyzed using a
nonlinear regression software to determine the kinetic equation parameters mentioned earlier.
Phase 3—Engineering Evaluation i
The research in Phase 3 focused on evaluation of three engineering issues: (1) chromium
speciation/separation; (2) corrosion; and (3) erosion. These issues were identified as critical
engineering design considerations for the SCWO process. !
Phase 4—Technology Development and Transfer !
The University of Texas at Austin (UT), College of Engineering, has a legislatively approved
and operational Center for Technology Development and Transfer (CTDT). The objective of the
Center is to dramatically reduce the time to move new ideas, laboratory developments, and designs
from the laboratory through the incubation period Into production.
i
i
4.2 Tasks and milestones
The project tasks and milestones are summarized in Figure 4.1. All tasks specified in the
proposal were accomplished. Some tasks were modified and other tasks were executed according to
a mc'dified schedule.
4.3 Quality assurance project plan (QAPP) \
The critical measurement in this work was the concentration of the selected compounds and
the oxidation products. Duplicate measurements were made to evaluate laboratory analytical
precision. The quality assurance (QA) objectives for the critical measurement Included precision,
accuracy, and completeness. A precision of less than ±10%, as expressed by Relative Percent
Difference (RPD), was achieved. Data that did not meet this criterion were rejected. Independently
prepared reference standards were used to evaluate accuracy. Accuracy of ±10% was achieved.
Completeness of 90% was achieved.
15
-------
I
I
1
t
i
1
I
1
T
L
t;
OS
V)
£
16
-------
The internal quality control (QC) check for sample analyses included the following: (a)
calibration check standards; (b) duplicate injections of single samples; (c) reference standards; and (d)
field blanks.
Both external and internal calibration methods were used for the quantification of acetic acid,
2,4-dichlorophenol, pentachlorophenol, pyridine, and 2,4-dichlorophenoxyacetic acid methyl ester in
influent and effluent samples. Calibration standards were prepared to span the: full range of
concentrations to be analyzed. If the relative standard deviation (RSD) of the calibration curve was less
than ±15%, the average response factor (RF) was used to determine the concentrations of the
experimental samples. If the RSD was not less than 15%, a calibration curve of concentration versus
area; counts was developed. Thus, instrument data were converted to sample concentrations.
A calibration check was made every 10 samples. If the RF of the calibration check was not
within ±15% of the RF of the initial calibration standards, all 10 samples analyzed; after the last
acceptable calibration check standard were reanalyzed. A reference standard was analyzed daily and a
percent bias of ±10% of the true value was accepted. A daily field blank accompanied experimental
samples. The purpose of the field blank was to detect possible contamination of samples. Distilled,
deionized water was used as the field blank.
4.4 Materials ;
Analytical grade acetic add, pentachlorophenol, 2,4-dichlorophenol, and pyridine were
obtained from Aldrich Chemical Co. and were used as received. 2,4-D methyl ester was obtained from
Sigma Chemical Co. Oxygen (99.5% purity) was suppfedl^iWDspn Oxygen. DisjilJ^raetori^BdjwafecL;
was used to prepare the feed solution and anafytk^ diiu&ns?^ ^ ^ " " "*"e:--';^*^. •-^^•--^^
Chemicals used for calibration or identification purposes included formic add (Fisher, 90%
aqueous), methanol (Maliinkrodt, 99.9%), dimethyiamine (anhydrous) (Kodak, reagent grade]!,
acetone (EM Science, reagent grade), sodium tetraborate (EM Science, reagent grade), and sodium
carbonate (anhydrous) (Spectrum, reagent grade). The following chemicals were obtained from
Aldrich: o-cresoi (99+%), acetic acid (99.7%), glycolic add (99%), propanoic add (99%), acrylic acid
(99%), maleic acid (99%), and L-lactfc add (sodium salt) (99%).
Calibration for off-gas analyses utilized two gas mixtures (Scotty Specialty Gases). Gas Mixture
A contained carbon dioxide (1.00% by vol.), carbon monoxide (1.03%), oxygen (1.0%)* and methane
(1.00%) in nitrogen. Gas Mixture B contained carbon dioxide (5.00%), carbon monoxide (5.0%),
carbon dioxide (1.00%), nitrogen (4.8%), oxygen (5.1%), and methane (4.0%) in helium.
Municipal and industrial wastewater sludges were used in the chromium speciation/separation
tests. The municipal sludge was an anaerobically digested sludge obtained from the City of Austin
Hoirnsby Bend Treatment Facility. The industrial sludge, an excess activated sludge, was collected at
17
-------
its source and shipped to Balcones Research Center, The University of Texas at Austin. These
sludges were stored at 4°C prior to use.
18
-------
5. EXPERIMENTAL PROCEDURES
This section describes the test setups, operating procedures, and sample analyses.
5.1 Test procedures
Four SCWO reactor systems were used in this project. The batch reactor system #1 and
continuous-flow reactor system #1 were used in organic destruction surveys and kinetic studies,
respectively. A bench-scale unit (continuous-flow reactor system #2) was used in the chromium
i
speciation/separation study. The batch reactor system #2 was used in the corrosion study. A brief
description of these test setups and operating procedures is provided below.
5.1.1 Batch system #1
As shown in Rgure 5.1, the batch reactor system #1 consisted of a reactor vessel assembly,
fluidized sand bath, wrist action shaker, water quench bath, temperature-pressure control panel, and
reactor vessel transfer assembly. The entire batch reactor system was housed in a walk-in fume hood.
The batch reactor vessel assembly, Rgure 5.2, was constructed using a U-shaped SS 316
tube fitted with a type J thermocouple and a shut-off valve. The inner and outer diameters of the tube
were 0.85 cm and 1.27 cm, respectively. The volume.of the reactor was 20 mL
•' ••":"•• "vw-SfeivS".'. - • ' - • . -".-'. x..••vWvsf:'?— -..;- '.'- '•• -. • .
In a typical batch experiment, the sand bath was first bleated to a desired temperature level.
For each test, the reactor was prepared as follows: (1) the reactor vessel was purged with oxygen; (2)
six mL of the feed stock was loaded; (3) the plugs and caps on both ends of the! reactor were
assembled; (4) oxygen was added and the reactor was isolated using the shut-off valve; and (5) the
reactor was inspected for leaks using a Snoop liquid leak detector. The reactor assembly was then
attached to the wrist action shaker. The reactor-shaker assembly was immersed in the fluidized sand
bath by using the pneumatic servos. Fluid temperature was monitored by a thermocouple connected
to the reactor head. The temperature readings were recorded manually every 15 sec. The reaction
temperature was maintained within ± 10°C of the desired temperature.
After a preselected reaction time, the reactor assembly was lifted from the sand bath and the
reaction was quenched in a water bath. Within 1 min the temperature was about 100°C, and ambient
conditions were reached after about 3 min. After ambient conditions were achieved, the reactor vessel
i
was removed from the water bath.
19
-------
Control
Switch
Oxygen
Horizontal Transfer Cylinder
Temperature
Recorder
Vertical
Transfer
Cylinder
Reactor Vessel
Assembly
Sand Bath
Air
Figure 5.1 Batch reactor system ft.
20
-------
Thermocouple
Shut-Off Valve
U-Shaped Vessel
Figure 5.2 Batch reactor assembly ft.
21
-------
5.1.2 Batch system #2
The design of the batch reactor system #2 was based on prior SCWO laboratory experience
and existing apparatus. This experimental apparatus consisted of three replacable straight-tube
reactors (each attached to a pressure gauge), a furnace (Thermolyne) with a temperature controller
(Fumatrol II), and a pressure generator (HIP) (Rgure 5.3). '
As shown in Figure 5.4, each reactor was made of a seamless metal tube 50.17 cm long with a
1.27 cm OD. The wall thickness of the SS reactors was 0.216 cm; volume was 28 mL. The Hastelloy
C-276 and Monel 400 reactors were fabricated using tubes with a wall thickness of 0.165 cm; volume
for each reactor was 29 mL. A Swagelok reducing union was used on one end and a cap or plug was
welded on the other end.
In each experiment the reactor was constructed of the same alloy as the coupons to minimize
contamination of the corrosion products and to prevent bi-metallic or galvanic corrosion. Three metal
samples (coupons) were inserted into each reactor. A SS 316 plug 24.5 cm long was inserted into the
open end of each reactor during the tests to minimize the volume of fluid trapped outside of the
isothermal zone of the reactor. A small hole was drilled through the top of the plugs in a diagoruil
manner, and this access hole was used to introduce oxygen into the reactor and as a means of
transmitting fluid pressure to the gauges.
Before each test, the coupons were inserted into the reactor, and a specific volume of the tesst
fluid was added. The volume was related to the desired fluid density. Since the test fltiid consisted erf
primarily water, fluid density at experimental conditions was determined from pressure-volurp©-
temperature relationships of water. During the experiment, temperature and pressure Were monitored
and adjusted to ensure the required fluid density.
Next, the reactor plug was inserted, and the reducing union located at the open end of the
reactor was tightened. Three reactors, each containing a different pH test fluid, were used
concurrently at each experimental temperature. Each reactor was connected to a pressure monitoring
assembly and charged with oxygen. Before placing the reactors in the furnace, all fittings and
connections were inspected visually for leaks using leak detection fluid. The furnace was set to the
specified experimental temperature. ,
The pressure gauges provided continuous data on pressure build up. It usually took 1 hr for
the reactors to reach thermal equilibrium and develop a stabilized pressure. If too little test fluid was
injected into a reactor prior to the start of the test, it was possible to manually adjust the pressure and
thereby maintain the required fluid density. If too much test fluid was added into ;a reactor, this
pressure was decreased by opening a pressure control valve and venting some of the fluid.
22
-------
FURNACE
OXYGEN CYLINDER
AND PRESSURE
REGULATOR
VALVES FOR
OXYGEN AND
PRESSURE
CONTROL
\ E
1)777771 *•
IS
K
I
a
fiTj/7
TEMPERATURE
CONTROLLER
ALL 0.0625 (I/16 in.)00
SS 316 TUBING
ROLLING CART
WITH REACTOR
RACK AND PRESSURE
GAUGES
-------
•=>
_J
Q.
O
uj
O
tft
C>
£
o
8
Nr^i
XI-
., SO. 17cm — '
REACTOR VESSEL (19 3/4 In.)
' '- - - pr
~> K7
5 | i
o £
|/
» 1
sSp'
H
7771
J 1
a
•6
f-
i
o
o
ui
— «n
-------
Pressure was increased by injecting deionized water using the pressure generator. This
device was connected to the reactor in the same manner as the oxygen system, this method of
increasing the pressure was used when only small (<3.45MPa) pressure increases were required.
When a pressure increase greater than 3.45 MPa was required, the test was aborted.
Typical coupon exposure time was 100 hrs. Throughout the test period the pressure was
monitored and adjusted as required. At the end of each test, the furnace doors were opened and the
cart was rolled out. The reactors were quenched with water. Each reactor was disconnected from the
pressure monitoring assembly. The coupons were removed in the same order in which they were
placed in the reactor. Liquid effluents were collected for pH analysis.
5.1.3 Continuous-flow system #1
As shown in Figure 5.5, the laboratory-scale, continuous-flow SCWO reactor system #1
consisted of a feed tank, feed pump, high pressure oxygen supply/control subsystem, preheater,
coilecl-tube reactor, fluidized sand bath, ice bath (effluent cooler), back pressure regulator, gas-liquid
separator, and an on-line off-gas analyzer.
The feed tank was a 20-gallon polyethylene container fitted with three Teflon tube
connections used for venting, helium purging, and feed supply. The feed solution was purged with
helium prior to each experiment to reduce the amount of dissolved oxygen and nitrogen. The feed
tank was mounted on a magnetic stirrer (Bel Art Products Cool Stir, Model 14-509-3). A 7.62 cm long
stir bar was used for mbdng. , ,-...-
A diaphragm metering pump (American Lewa, Model HLM-1) was used to pressurize the feed
solution. The throughput of this pump was rated 150 mL at 69 MPa. The pump was calibrated
frequently with water at the test pressure by measuring the effluent volume over a selected time
interval.
The pressurized feed solution was preheated prior to entering the sand bath. Two radiant
heaters (Watlow, Model 9224C, 850 watt each) were used to provide heat to the preheating section
comprising 1.83 meters long, 12.7 mm O.D., and 2.1 mm wall thickness SS 316 tubing. The radiant
heaters were controlled using two 120 volt variacs. The feed solution leaving the preheater flowed
into an additional 6.07 meters of 6.35 mm O.D. SS 316 tubing arranged in the fluidized sand bath
(Techne, Model SBL-2D). The preheated feed solution then flowed into a Swagelok 6.35 mm (0.25
in) SS 316 cross located immediately above the fluidized sand surface. This arrangement served as a
mixer. The reactor inlet was attached to the bottom port of the mixing cross. The coiled tube reactor
was made of 6.35 mm O.D. and 0.89 mm wall thickness Hastelloy C-276 tubing. Several reactors were
made using different tubing lengths (2.13, 3.96, and 6.07 meters) and these provided variable
residence times. The variation of the reactor length allowed the flow rate and
25
-------
r
Filter
Mnnifold
-------
Reynolds number to remain constant for a given temperature. Reynolds number was controlled at
about 8000 to ensure the ideal plug-flow reactor assumption (Bird et a!., 1960). .
A thermocouple (type K with Inconel 600 sheath, Omega, P/N KQIN) was inserted at the top
port of the mixing cross and then into the reactor to determine the fluid temperature in jthe reactor inlet
section. A similar thermocouple was inserted in the effluent end of the reactor. The average of these
two fluid temperatures was taken as the reaction temperature.
The reactor effluent was cooled by a heat exchanger consisting of an SS 316 coiled tube
(6.07 meter long, 6.35 mm O.D.), which was submerged in an ice bucket. The cooled effluent then
passsed through a 0.5 micron SS 316 filter and then to the back pressure regulator (Tescom, Model 54-
2122W24). This cooled and low-pressure effluent then flowed into a gas-liquid separator consisting of
a 30.5-cm high, 10.16-cm O.D. Lexan cylinder filled with 1.43-cm diameter glass pellets. Gaseous
"'effluent was removed from the top of the separator and directed to the vent. A series of valves allowed
the gaseous effluent to be re-directed through a Fisher Hamilton gas part'rtioner for on-line off-gas
analyses. The liquid effluent flowed from the separator into a container for disposal.! Liquid effluent
samples were collected from a sample port located in the effluent line from the gas-liquid separator.
Oxygen was pressurized by an air-driven booster (Haskel, Model 27267), which elevated the
pressure from about 3.5 MPa (500 psi), as supplied from bottled oxygen, to 31 MPa (4500 psi). The
oxygen flow rate was controlled by a metering valve (Badger Research Control, Model
1001GHT3635V) fitted with a P-14 trim. The oxygen mass flow rate was digitally displayed using a
mass flow meter assembly (Brooks, Model 5860). The oxygen volumeric flow rate was calibrated using
a soap bubble rneter.
5.1.,4 Continuous-flow system #2
The continuous-flow reactor system #2 was an existing pilot-scale unit comprising a high-
pressure pump, a concentric-tube reactor, nine electric band heaters, thermocouples, pressure
gauge/transducer, weighing scales, control/data acquisition equipment, insulation, and safety shields.
The reactor consisted of two concentric SS 316 tubes. The outer tube dimensions were 5.08
cm O.D., 2.54 cm I.D., and 5.74 m length. The inner reactor tube dimensions were 0.95 cm O.D., 0.62
cm I.D., and 5.70 m length. Total fluid volume was 2673 mL Since the oxidant injection port was
located at the middle point of the downcomer, the bottom half of the vessel was considered the
reaction zone and had a volume of 1337 mL. ;
Influent entered the top of the reactor and flowed downward through the annular section,
while the effluent flowed up through the core section. The reaction zone of the reactor consisted of
the lower 2.7 meters of the annular section of the reactor. Sample ports were installed at the top and
bottom of the reactor zone. '
27 ;
-------
External heat was provided to the reactor by nine 1000-watt electric band heaters (AKINSUN
Model EXB-200600). These heaters were distributed along the length of the outer reactor wall. The
temperature of station G located inside the core of the reactor was used as a feedback to a temperature
controller (Eurotherm Model 847). The controller operated the six lower band heaters. The top three
heaters were operated manually. Six type K, ungrounded thermocouples were attached onto the
reactor surface. Two type K sheathed thermocouples were located within the reactor to monitor fluid
temperatures. Temperatures could be monitored at a total of eight different locations.
The surface of the reactor was covered with three layers of thermal insulation. The first layer
was Kaowool ceramic fiber (2.5 cm thick). The second layer consisted of a 2.5 cm thick ridged Kaylo
pipe insulation. Finally, an additional layer of Kaowool ceramic fiber was installed around the bottom
half of the reactor. I
The influent liquid was pumped with an American Lewa Diaphragm Pump, Model HLM-1, with a
pump capacity of 130 g/min. Fluids in the effluent line and sampling lines were cooled by chilled water
using a double-pipe heat exchanger. Pressure in the reactor was regulated by a metering valve and
monitored by both an Ashcroft pressure gauge and a Heise pressure transducer interfaced with a
digital display (Beckman Industrial 600 Series). The weight of the influent tank, as determined by a
scale (Ohaus Model PBI-01 -920-44), was recorded to calculate the mass flow rate.
The oxygen pressure was increased to 26.5 MPa using a gas booster (Haskel Model 27267).
A Foxboro D/P Cell and a Badger metering valve were used to mpnftor and regulate the oxygen mass
flow irate, respectively. The capacity of the booster output was 3 kg oxygen per hr. The high-pressure
oxygen was transferred through a 0.32 mm O.D. Monef 400 oxygen tubing. No additional oxygen
mixing device was provided within the reactor. Additional Information aboirt th^ continuous-flow"
reactor system #2 and typical test results can be found elsewhere (U et al.f 1990).
S.2 Feed preparation
For the organic destruction and kinetic studies, three aqueous feed solutions were prepared
for each selected compound. The concentration of these feed stocks varied from 45 mg/L to 3000
mg/L depending on the solubility of the compound in water at ambient conditions. Test solutions were
prepared directly from acetic acid, pyridine, 2,4-dichlorophenol, and 2,4-D methyl ester. Since the
solubility of pentachlorophenol was only 14 mg/L, methanol was used as a cosolvent. !
For the corrosion study, several solutions were made: (1) a solution with a pH of 2.1 was made
using a mixture of hydrochloric acid (HCI) and de-ionized water, (2) a solution with a pH of 5.8 was made
from a mixture of sodium chloride (NaCI) and de-ionized water; and (3) a solution with a pH of 8.6 was
made from a mixture of sodium chloride, sodium hydroxide (NaOH), and deionized water. In the
28
-------
solutions with pH values of 5.8 and 8.6, enough sodium chloride was added to achieve the same
chloride concentrations as the solution with a pH of 2.1, which included hydrochloric acid.
Municipal and industrial wastewater sludges were used as feed streams in the chromium
speciation/separation study. The municipal sludge was an anaerobically digested sludge obtained
from the City of Austin Homsby Bend Treatment Facility. The industrial sludge, an excess activated
sludge, was collected at its source and shipped to the Balcones Research Center, The University of
Texas at Austin. These sludges were stored at 4°C prior to use.
5.3 Sample collection and analyses
Liquid, gaseous, and solid effluent samples were collected and analyzed. A
sample collection procedures and analytical techniques is provided as follows:
description of
5.3.1 Sample collection
Liquid samples were collected in a 35-mL sample vial (Teflon-lined, siiicone septum-seal). To
meet the QA objective for representativeness, only those samples for which recoveries were greater
than 90% of the initial volume were accepted. Immediately after collection and labeling, each sample
was stored at 4°C. Most of the samples were analyzed within 24 hrs. However, these compounds
appeared to be relatively stable at 4°C.
Gaseous samples were drawn from the gas-liquid separator (continuous-flow reactor system
#1) and directly injected Into a nearby gas chromatograph for real-time analyses.
5.3.2 Gas ehromatography (GC)
The GC technique was used to analyze both liquid and gaseous effluent samples. The first GC
(Hewlett Packard, Model 5890) equipped with both Flame lonization Detector (FID) and Electron
Capture Detector (ECD) was used to determine the concentration of 2,4-dichlorophenol, pyridine, and
2,4-dichlorophenoxyacetic acid methyl ester. A fused capillary column (Supeteo, SPB-608) was used
to analyze the chlorinated phenolic compounds. Pyridine samples were analyzed using a different
column (Supelco DB-WAX) and a FID detector. In this case the minimum detection limit for the pyridine
was about 1 mg/L. The ECD detector was used for 2,4-dichlorophenol and 2.4-dichlorophenoxyacetic
acid (2,4-D) because of its superb sensitivity for chlorinated compounds. The minimum detection limits
of the ECD were 0.03 mg/L and 0.1 mg/L for 2,4-dichlorophenol and 2,4-D methyl ester.
The second GC—a gas partitioner (Fisher Hamilton, Model 29) equipped with two packed
columns, a thermal conductivity detector (TCD), and an integrator (Hewlett Packard, Model 3392A)—
was used for the off-gas analysis. The first column was a 0.6 meter long and 6.35 mm O.D. SS 31(3
tube packed with silica gel (Supelco Davidson 923,100/200 mesh). This column separated the CO;>
29 i •
-------
from the other gases. The second column was a 1.98 meter long and 3.18 mm (0.125 in) O.D. SS 316
tube packed with a molecular sieve (Supeico 5A, 45/60 mesh). This column separated the Oa, Nj>,
CH4, and CO. The GC was operated isothermally at 25°C. Helium flow rate was 20 mL/min. The gas
partitloner was calibrated using prescribed gas mixtures as explained in Section 4 on Materials and
Methods. Averaged deviations for CO2, CO, and CH4 were less than 4.5%, 4.4%, and 3.4%,
respectively.
External standards were used for the calibration of all GC analyses. Calibration concentrations
ranged from 0.01 mg/L to 300 mg/L. At least 10 standard solutions were injected to verify the linearity
of the response. Three injections were made for each standard solution and sample. The
quantification of the response was based on area counts under the peaks. A Hewlett Packard 9000
computer, connected to the gas chromatograph system, was used to perform the computation. For
most samples, a precision of ±5% was achieved. Where this level of precision was not achievable, a
precision of ±10% was accepted.
Reference standards were used to evaluate the accuracy of the calibration standards and the
analytical procedures. Daily analyses were made and a percent bias of ±10% of the true value was
accepted.
5.3.3 GC-mass spectroscopy (GC-MS)
The unknown peaks found in the HP GC analyses were further Identified using a second gas
chromatograph (Varian, Model 3400). This GC was equipped with amass spectroscope, an iort trap
detector (Rnnigan Mat 700), and a SGE BP5 column. The temperature program begari at 35°C, held
for two min, ramped at 10°C/min to 170°C, and malntaJried for 1 min.
5.3.4 Ion chromatography (1C)
An ion chromatograph (Dionex, System 14 1C) equipped with a conductivity meter and a
separation column (Dionex, HPIC-AS1) was used in the analysis of acetic acid. The eluent was a
mixture of sodium bicarbonate (0.003 mole/L) and sodium carbonate(0.0024 mole/L) in distilled and
deionized water. The eluent solution eliminated the base line interference. Standard concentrations
ranged from 10 mg/L to 100 mg/L The pH of the samples and standards were maintained between 7
and 8. All 1C analyses were performed at room temperature. ;
5.3.5 pH
An Orion Model SA 720 pH meter was used. The meter was standardized using Fisher
Scientific pH standards (pH 4,7, and 10).
30 !
-------
5.3.6 Chemical oxygen demand ,
Chemical oxygen demand (COD) analyses were performed on samples containing excess
activated sludge and anaerobically treated sludge. Dichromate, in a sulfuric acid solution, was used as
the oxidizing agent. Standard Method 5220 D (Clesceri et al., 1989) was used. '
5.3.7 Ammonia
Ammonia concentration was determined using an Orion Research Model 701A digital
ionalyzer. This device was equipped with an ammonia specific probe. The meter was standardized
using ammonia standards (Orion Application Solution -1 mg/L, 10 mg/L, 100 mg/L, and 1000 mg/L).
5.3.8 Total suspended solids
Total suspended solids (TSS) concentrations were determined by Standard Method 2540 D
(Clesceri et al., 1989). A known volume of representative sample was filtered through a tared 0.45 urn
filter. The fitter paper was dried to a constant weight. Final weight of the filter minus the tare weight was
used to determine the TSS concentration in mg/L.
3.3.9 Liquid chromatography
Pentachlorophenol (PCP) was analyzed using a high performance liquid chromatograph
(HPL.C). As compared to the GC.technique, HPLC analyses saved considerable time. The HPLC
system was equipped with an SS column (Supelco L- PAH, 150 mm x 4.6 mm, particle size Sum), and
a UV detector set at 313 nm. vfhe>4EC software program MAXIMUM vwas used f^r idata acquisition and
processing. The mobile phase was a mixture of 60 parts acetonftrile, 25 parts methanol, and 15 parts
water. Flow rate and pressure were 2.0.mL/min and 15.2 MPa, respectively. The injection volumes
ranged from 20 \iL to 200 \iL. Standard curves for PCP were generated covering a range of 0.04 mg/L
to 10 mg/L. The retention time of PCP under these HPLC conditions was about 1.19 mfn.
Standard solutions of pentachlorophenol were prepared by diluting a 10 mg/L stock solution.
This stock was prepared by dissolving 10.010 mg PCP in one liter of warm (50°C) distilled and
deioriized water. The minimum detection limit for PCP by this method was 0.04 mg/L. The solubility of
pentachlorophenol in water at 25°C was 14.0 mg/L.
3.3.10 Chromium species
The total soluble chromium concentration was defined as the sum of the hexavalent chromium
and soluble trivalent chromium. Sample filtration and storage for total soluble chromium and
31
-------
hexavalent chromium analyses were performed according to Standard Method 3030 B (Clesceri et a!.,
1989). :
Hexavalent chromium was determined by Standard Method 3500-Cr D (CJesceri et al., 1989).
This method utilized the absorption, at 540 u.m, of the hexavalent-diphenylcarbazone complex.
Absorbance was measured using a Bausch and Lomb spectrophotometer with a 1-cm cuvette. The
detection limit was 0.004 mg/L. Analyses were performed by preparing a standard curve of known
hexavalent chromium concentration. Absorption was plotted against concentration. Hexavalent
chromium concentration was determined by comparing the absorption of the sample against the
standard curve.
Total soluble chromium was determined by Standard Method 3111 (Clesceri et al., 1989).
Light absorption was set at 362.7 um The flame atomic absorption measurements were performed
with a Perkin-Elmer Model 303, single-head spectrophotometer using air as the auxiliary oxidant arid
acetylene as fuel. The detection limit was 0.01 mg/L. The analyses were performed by preparing a
standard curve of known soluble chromium concentration.
Soluble trivalent chromium concentration was determined by subtracting the hexavatent
chromium concentration of a sample from the total soluble chromium concentration of the sample.
Samples for total chromium determination were prepared by Standard Method 3030 F
(Clesceri et al., 1989). A known volume of digested sample was analyzed for total soluble chromium by
fJame atomic absorption;. , \
Insoluble trivalent chromium concentration of samples was determined by subtracting
soluble trivafent chromium concentration from the total chromium concentration.
. - •--•• - -• - -• ' " "' ' '
32
-------
6. RESULTS AND DISCUSSION _
Results were obtained from the batch, continuous-flow, corrosion, chromium
speclation/separation, and model development studies. Detailed information may be obtained from
the various theses and presentations as listed in Appendix A. Typical results and data quality are
presented below. :
6.1 Data quality
Table 6.1 summarizes the data quality in terms of precision, accuracy, and completeness for all
reported parameters. The tabulated values represent either upper or lower limits of the data quality
indicators. Tables 6.2 and 6.3 provide sample analytical reports for 1C and GC measurements,
respectively, In addition, chromatographic analyses, calibration methods, and detection limits for evety
compound quantitatively studied are specified in Table 6.4.
6.2 Batch test
The purpose of batch tests was to provide treatability data. Relative destruction; characteristics
-' r -""'*• 'r " ,<*-_-, • ^ _ - ^ ^ . .^ ^g >;,v h -___ .*,.„ ;^ ,.,£:-,-- i^r*-ft i|j^£/.- ,^ ( • '
of the selected model compounds were established. These tests utilized the apparatus described in
'•" . "' :.- -", '-";.' '."'.• • r- "•.,»'""..', .'..-'.••. .. - ' - -. ''<£'. y> '- -'. -.-V'Kf:! •^.'W-'-^V'.if *•?•' ';':""'
Section 5.1.1. The compounds evaluated consisted of acetic add, pentachforoprienol(PGP),£4-
dichlorophenol (2,4-DCP), pyridine, and 2,4-dichIorophenoxyacetlc acid (2,4-D) methyl ester. The
batch test conditions are summarized in Table 6.5. !
6.2.1 Acetic acid
A total of 35 acetic acid tests were performed (Table 6.6.) Reaction temperature and pressure
ranged from 378°C to 513°C and from 241 bar to 345 bar, respectively. Effluent samples were
analyzed by 1C. The concentrations of acetic acid feed solutions were 500 mg/L to 3000 mg/L, and the
retention times ranged from 1 to 20 min. Oxygen loadings ranged from 130% to about 300% of the
stoichiometric demand. The destruction efficiencies increased with reaction temperatures, from 20%
at 400"C to 99% at 490°C. A global kinetic model was developed for acetic acid based on the batch
test results. The activation energy for the oxidation of acetic acid was estimated to be 110 kJ/mole.
33
-------
TABLE 6.1 SUMMARY OF DATA QUALITY
Reported Parameter
Temperature
Pressure
Pressure
Time
Weight
Volume
Volume
Liquid Flow Rate
Gas Flow Rate
Feed Concentration
Effluent Concentration
Effluent Concentration
PH
Ammonia
Total Suspended Solids
Hexavalent Chromium
Total Soluble Chromium
Total Chromium
Measurement Technique
Thermocouple
Gauge
Transducer
Stop Watch
Balance
Volumetric Rask
Pipette
Calculation
Flowmeter
Calculation
GC.IC.HPLC
COD
Electrode
Electrode
Balance
UV-VIS Absorption
Atomic Absorption
Atomic Absorption
Precision
±2
±3
±3
±0.5
±0.1
±1
±1
±10
±10
±10
±10
±10
±10
±10
±3
±10
±10
±10
Accuracy
±2*C
±0.1 MPa
± 0.1 MPa
± 0.1 sec
±0.1 mg
± 0.2% FS
± 0.2% FS
± 0.5 g/min
± 0.5 cc/mii
± 0.2 mg/L
± 10%
±10%
±0.05
±10%
±1%
±10%
±10%
±10%
Completeness
100
100
100
i
100
I
100
100
100
160
1 100
100
>90
>90
>90
>?o
>90
>90
>90
>90
FS = Full Scate
6.2.2 Pentachlorophenol (PCP)
A total of 53 pentachlorophenol destruction tests were performed (Table 6.7). Reaction
temperatures and pressures, respectively, ranged from 400°C to 500°C and from 241 bar to 345 bar.
Sincei the solubility of pentachlorophenol in water at ambient conditions was low, 14 mg/L, a special
injection technique was used. A concentrated solution of pentachlorophenol in methanol (100
34
-------
TABLE 6.2 SAMPLE ANALYTICAL REPORT FOR 1C MEASUREMENTS
Target Compound
Acetic Acid (Aqueous Solution)
Test Date: 1/15/91
Analysis Date: 1/17/91
1C Operating Condition^
Column: AS1
Sample Inj. Vol.: 2 cc (Int. Sample Loop = 100 j
Eluent Flow Rate: 2 cc/min
Eluent: 0.003 M NaHNOa
+0.0024 M Na2NO3
aqueous solution
Concentration Calibration
Detector:
Detector Gain:
System Pressure:
Temperature:
Chart Speed:
Printer Gain:
Conductivity
2.8-3.5 MPa
20'C
0.635/min
1.0
Standard Concentration
(mglti
30.0
50.0
70.0
100
Peak Height
57.0
97.0
135
194
Response Factor
(mnVal)
1.90
1.94
1.93
1.92
Calculated Concentration
(mg/L)
29.8
50.3
69.8
100
- — — » +*-r t u^ffm I \J\J [
Linear regression: Concentration (mg/L) = 0.512xPeak Height (mm) + 0.655 r2 = 0.9999
Precision !
Sample ID No. Dilution Ratio 1st Inj. Peak Height 2nd Inj. Peak Height RPD
— talS)! (mm) (%)
0115-1
1:5
129 130 0.77
Accuracy
Tirue Concentration
(mA.)
Peak Height Measured Concentration Bias
ferns) (mgi.) (%) !
66.2
130
67.2
1.5
35
-------
TABLE 6.3 SAMPLE ANALYTICAL REPORT FOR GC MEASUREMENTS
Target Compound
Pyridine (Aqueous Solution)
GC Operating Condition^
Column: SPB-5
Helium Row Rate: 14 cc/min
Sample Inj. Vol.: 1 jiL (splitless)
Run Program: Hold at 35"C for 4 min;
ramp at a rate of10'C/min to 100'C; and
ramp at a rate of 190"C/min to 190*C.
Detector FID
Detector Temperature: 250*C
Injector Temperature: 120"C
Concentration Calibration
Pyridine Standard
10.0
20.0
50.0
100
250
500
1000
Mean
Standard Deviation
RSD(%)
o-Cresol
Peak Area
25720
26008
24000
27529
25865
21936
24980
•>', .v—:" "* , ^-'
Pyridine
Peak Area
2888
5632
11961
27087
62614
105379
234350
.;"&V'.U . ."i" ••: • .
Response
Factor (RF)
0.2246
0.2165
0.1994
0.1968
0.1937
0.1922
0.1876
0.19T7
0.0137
Motes: The internal standard (o-cresol) concentration in each pyridine standard solution was 20 rna/IL
--;>•:-*• t^-*-* • . .^----,-.. •• - -• r-^fM^,?^^!^^^; •:.:_• -- -^;:>^f»Ss#.?:;:. .••'.•- "--
which was obtained by adding 100fiL of 2000 mg/Lo-cresol aqueous solution into 10 mL of each '
pyridine standard solution.
Precision
Sample ID No.
535
600
528
534
Dilution Ratio
1:4
. 1:4
1:4
1:4
1st Injection
Peak Area
21524
117758
2nd Injection
Peak Area
23011
116625
RPD
<%>
6.68
O.968
Notes: One of the internal quality controls used in this project involved a sample duplication
procedure. Immediately after collecting a sample (either feed or effluent), the operator split the sample
into two sample vials and labeled them with nonconsecutive ID numbers. These duplicate samples
were then analyzed by analytical staff, who were unaware that the samples were duplicates. In the
above example, Sample Nos. 535 and 600 are duplicate samples, as are Nos. 528 and 534.
36
-------
TABLE 6.3 SAMPLE ANALYTICAL REPORT FOR GC MEASUREMENTS, cont.
Accuracy
True Concentration Peak Area Measured Concentration Bias
fmg/L)
1960
1960
1177
1177
115596
102272
55733
60345
vy-j
1948
1941
1159
1187
US*
-0.61
-0.97
-1.5
+0.85
g/L) was used. Prior to injection of the pentachlorophenol solution, six mL of water was added to the
batch reactor. An aliquot of the pentachlorophenol solution, equivalent to the initial concentration of
100 mg/L to 500 mg/L PCP, was injected into the batch reactor. The reaction times ranged from 2 min
to 15 min. Oxygen loadings ranged from 130% to 200% of the stoichiometric demand. The HPL.C
method was used for pentachlorophenol analyses. '. •
The destruction efficiencies ranged from 99.70% to >99.99%. Reaction temperatures and
reaction times, respectively, varied from 400°C to 500°C, and 2 min to 15 min. Most of the final
concentrations of pentachlorophenol were below the detection limit of 0,04 mg/L. Severe corrosion
was observed in the reactors because one of the major by-products was hydrochloric acid. The
combination of chloride ion, oxygen, and water at these SCWO conditions was simulated In the
corrosion tests as described In Section 6.4.
rf;afe,%- ' ••^'-•- ••'••J''<•"•
mL of methane! wa^
\
the PCP concentration in the rinse solution was less than 0.6 mg/L, representing a 0.2% change in
PCP destruction efficiency based on an initial concentration of 500 mg/L
6.2.3 2,4-DlchIorophenol (2,4-DCP) ;
A total of 56 batch tests were performed on 2,4-DCP, including 27 scheduled tests and 29
additional tests at various intermediate conditions (Table 6.8). Reaction temperatures and residents
times, respectively, ranged from 400°C to 500°C and from 2 min to 15 min. The feed concentrations
used were 300 mg/L, 500 mg/L, and 1000 mg/L. Excess oxygen, 300% to 700% of the stoichiometric
demand, was provided.
Destruction efficiencies ranged from 39.93% to 99.99%. The destruction rate was moderate as
compared to PCP at the higher temperatures and acetic acid at the lower temperatures. The 2,4-DCP
tests, subject to operating conditions, exhibited 4 to 30 additional GC peaks, reflecting the existence
of transition products. The pH of effluent samples was about 2.5 due to the formation of
37 :
-------
TABLE 6.4 SUMMARY OF CHROMATOGRAPHIC ANALYSES, CALIBRATION
METHODS, AND DETECTION LIMITS
Target Analytical Calibration Reference Detection Standard
Compound (TComp) Technique Method Standard Limit Concentration
(Detector/Column) (mg/L) (rng/L)
Batch Tests
2,4-DCP
GC/ECD/SPB-5 External TComp 0.02
Pyridine
GC/FID/DB-WAX External TComp 0.5
PCP HPLC/UV/L-PAH External TComp
2,4-D methyl ester GC/ECD/SPB-608 External TComp
Acetic acid IC/Conductivity/AS1 External TComp
Continuous-Flow Tests
2,4-DCP GC/FID/SPB-5 Internal o-Cresol
Pyridine
GC/RDySPB-5
Internal o-Cresol
0.04
0.01
1
0.5
0.5
Low range:
0.03, 0.05, 0.1,
0.3,0.5,0.8, 1,
3,5, 8,10
High range:
10, 30, 50, 80,100
1,3,5,8,10,30,
50, 80, 100, 200,
300, 400, 500
0.1, 0.5, 1, 2, 5, 10
0.02,0.05,0.1,0.2,
0.4, 0.8,1,2, 5, 8,
10,20, 30,40
30, 50, 70, 100
1,10,25,50,
100, 250, 500
10, 20,50, 100,
250,
Phenol ,
• -' •£-,.,
Other chlorinated
phenols
Formic acid
Acetic acid
Maleic acid
Chloride
Nitrate
GC/RB/SjPBr5
GC/Flb/SP!B-5
IC/Conductivity/AS3
IC/Conductlvity/AS3
IC/Conductivity/AS3
IC/Conductivity/AS3
IC/Conductivtty/AS3
Internal
internal
External
External
External
External
External
o-Grejsol
o-Cresol
TComp
TComp
TComp
Naa
NaNOs
0.5
0.5
1
1
0.5
1
r-',.1-
1,
1,
5.
5,
1,
5,
2$. 5«~ ^0* 25 " * > '
2,5^10^25
2, 3. 4,
10, 15,
10, 20,
2.3.4,
10, 15.
5
20, 25
30,40
5
;20, 25
Notes: Detection limits were derived from visual observation of instrumentation responses.
38
-------
TABLE 6.6 ACETIC ACID BATCH TEST RESULTS
Experiment
Number
0108-3
0104-3
1226-2
1226-3
1226-5
0104-6
0108-6
0117-2
1227-7
1227-4
0117-3
0115-3
1226-1
0104-2
0115-4
1226-6
1226-7
0104-5
0104-7
1227-2
0108-7
0104-8
1227-6
0115-3
0115-1
0104-1
0104-4
0108-5
0117-1
Reaction
Time
(min)
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
Test
Temperature
(°a
397
401
419
419
431
450
462
467
494
498
498
396
396
405
420
434
445
446
465
477
480
483
513
395
399
402 ,
441
477
490
Influent
Concentration
(rng/L)
500
1500
3000
3000
3000
1500
500
3000
1500
3000
500
500
3000
1500
500
3000
3000
1500
1500
3000
500
1500
1500
3000
500
1500
1500
500
3000
Effluent
Concentration
(mg/L)
437
1190
2390
2500
2130
742
328
494
354
72.0
12.0
363
2470
1130
377
1150
1540
677
105
556
159
367
393
1730
338
- 1090
".- :^::£95>4,V
ioi ^
18.0
Destruction
Efficiency
'(%)
12.6
20.6
20.4
16.7
29.2
50.5
34.4
83.5
76.4
97.6
97.6
27.4
17.6
24.7
24.6
61.5
48.8
54.9
93.0
81.5
68.2
75.5
73.8
42.5
32.4
27.7
%m^JS7.7 •'. • -"•-T.--1
"*79.8" "
99.4
aqueous solution. This procedure required adding a given amount of 2,4-D to 500 mL distilled and
deionized water, heating the mixture up to about 75°C, and keeping the mixture stirred at this
temperature until the solid (2,4-D) disappeared. It required about two hrs to dissolve 850 mg of 2,4-D
in a one-liter aqueous solution. This procedure was verified by comparing the HPLC results. In this
case, a 50 mg/L 2,4-D reference solution was made at room temperature. The HPLC results and those
obtained from the 850 mg/L 2,4-D solution were within 4% RPD. i
The available analytical techniques for 2,4-D analysis presented a problem. The GC technique as
described in EPA 8150 Method required large sample volumes (1000 mL) and 5 hr of analysis time.
Fortunately, 2,4-D was successfully analyzed using the HPLC technique for concentrations as low as 2
mg/L (Oh and Tuovinen, 1990). In this case, an Altex model 100A HPLC was used with a Spherisorb
39
-------
TABLE 6.6 ACETIC ACID BATCH TEST RESULTS
Experiment
Number
0108-3
0104-3
1226-2
1226-3
1226-5
0104-6
0108-6
0117-2
1227-7
1227-4
0117-3
0115-3
1226-1
0104-2
0115-4
1226-6
1226-7
0104-5
0104-7
1227r-2
0108-7
0104-8
1227p-6
0115-3
0115-1
0104-1
0104-4
0108-5 . "
0117--1
Reaction
Time
(min)
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
10
10
10
10
10
10
10
20
20'
20
20
20
20
Test
Temperature
397
401
419
419
431
450
462
467
494
498
498
396
396
405
420
434
445
446
465
477
480
483
513
395
399
402
441
477
490
Influent
Concentration
(mg/L)
500
1500
3000
3000
3000
1500
500
3000
1500
3000
500
500
3000
1500
500
3000
3000
1500
1500
3000
500
1500
1500
3000
500 -Sr-
1500
1500
SOQf '
3000
Effluent
Concentration
(ma/L)
437
1190
2390
2500
2130
742
328
494
354
72.0
12.0
363
2470
1130
377
1150
1540
677
105
556
159
367
393
1730
1090
""" toi'""^"'^'
18.0
Destruction
Efficiency
12.6
20.6
20.4
16.7
29.2
50.5
34.4
83.5
76.4
97.6
i * •**
97.6
27.4
17.6
24.7
24.6
61.5
48.8
54.9
93.0
81.5
68.2
75.5
73.8
42.5
- 32 4 -' " •""'
—, ' ; -~-TT ^ -
••", u.v.tf7'"7' • ' ""*""' ,*•
j" '**? ^.CiXf '"*•* ••">-."•''''•"'*' • "'^^
sS!^Hillfe:^|
99.4
aqueous solution. This procedure required adding a given amount of 2,4-D to 500 mL distilled and
deionized water, heating the mixture up to about 75°C, and keeping the mixture!stirred at thts
temperature until the solid (2,4-D) disappeared. It required about two hrs to dissolve 850 mg of 2,4-D
in a one-liter aqueous solution. This procedure was verified by comparing the HPLC results. In this
case, a 50 mg/L 2,4-D reference solution was made at room temperature. The HPLC results and those
obtained from the 850 mg/L 2,4-D solution were within 4% RPD.
The available analytical techniques for 2,4-D analysis presented a problem. The GC technique as
described in EPA 8150 Method required large sample volumes (1000 mL) and 5 hr of analysis time.
Fortunately, 2,4-D was successfully analyzed using the HPLC technique for concentrations as low as 2
mg/L (Oh and Tuovinen, 1990). In this case, an Altex model 100A HPLC was used with a Spherisorb
40
-------
TABLE 6.7 PENTACHLOROPHENOL BATCH TEST RESULTS
Experiment
Number
051891-1
052091-1
052091-2
052091-5
052991 -A
052191-1
052891-4
053091-3
052091-7
052991-3
053191-1
051791-2
051791-3
051791-4
052091-3
052091-6
052991 -B
041691-1
041691-2
041691-3
052891-3
053091-3
042391-1
042391-2
042391-3
052091-8
052991-2
051591-1
051591-2
051591-3
052091-4
053191-4
041691-4
041691-5
041691-6
052891-1
052891-2
053091-1
042391-4
042391-5
042391-6
052091-9
053191-3
052991-1
051691-1
051691-2
051691-3
041891-1
041891-2
041891-3
042391-7
042391-8
042391-9
Reaction
Time
(min)
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
4
.4
4
4
4
4
4
4
4
4
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
15
15
15
15
15
15
Test
Temperature
400
400
400
400
400
425
450
450
500
500
500
500
500
500
400
400
400
400
400
400
450
450
450
450
450
500
500
500
500
500
~' "400
400
400
400
400
450
450
450
450
450
450
500
500
500
500
500
500
400
400
400
450
450
450
Influent
Concentration
(mg/L)
100
100
100
100
300
100
100
300
100
300
300
500
500
500
100
100
300
500
500
500
100
300
500
500
500
100
300
500
500
500
100
300
500
500
500
100
100
300
500
500
500
100
300
300
500
500
500
500
500
500
500
500
500
Effluent
Concentration
(mg/L)
BDL
BDL
BDL
BDL
BDL
BDL
BDL
0.72
BDL
BDL
0.87
0.93
BDL
0.93
BDL
BDL
BDL
BDL
0.72
BDL
BDL
0.51 -
0.95 "" *
1.4 .
BDL*^8^'
0.42
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Destruction
Efficiency
(%)
>99.96
>99.96
>99.96
>99.96
>99.99
>99.96
>99.96
99.76
>99.96
>99.99
>99.71
99.81
>99.99
99.81
>99.96
>99.96
>99.99
>99.99
>99.99
>99.99
>99.96
99.76
>99.99
>99.99
>99.99
>99.96
>99.99
•;" J99.90 ,,
i,jJ99V72-»l-^
'" '^>§!j'9Q^G
>99!e6 " *
>99i99vi
>99'99^
>99.99
>99.96
>99.96
>99.99
>99.99
>99.99
>99.99
>99.96
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
: Reactor
| Rinse
i (mg/L)
j
,
0.28
*
;• BDL
'
0.78
0.91
I
i
',
(
0.78
0.81
;, 0.67
0.75
!
[
'
[
1
1.11
!
!
BDL: Below detection limit (0.04 mg/L)
41
-------
TABLE 6.8 2,4-DICHLOROPHENOL BATCH TEST RESULTS
Experiment
Number
0710-1
0715-2
0702-3
0715-1
0702-1
0717-4
0711-3
0716-2
0711-2
0708-3
0716-8
0712-4
0701-3
0705-1
0712-1
0711-1
0703-1
0710-2
0715-4
0715-3
0702-2
0705-2
0711-4
0716-3
0708-4
0716-7
0705-3
0705-3A
0712-3
0717-3
0628-1
0708-2
0701-1
0716-5
0710-3
0709-2
0715-5
0708-5
0709-1
0709-6
0711-5
0716-4
0711-6
0716-6
0717-2
0717-1
0705-5
Reaction
Time
(min)
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
5
5
5 .
5
5
5
5
5
5
. 5 ?;. ..
5
5
5
5
5
5
5
7
10
10
10
10
10
10
10
10
10
10
10
10
10
Test
Temperature
394
396
398
398
412
441
446
446
448
462
471
496
500
503
517
429
362
399
402
403
408
429
443
447
450
U70
"•"" 487
488
493
494
507>
539
550
451
390
396
405
430
432
435
446
446
448
469
491
499
499
influent
Concentration
(man.}
300
1000
500
1000
500
500
300
1000
300
500
1000
300
500
500
300
300
500
300
1000
1000
500
500
300
1000 .
500,
'-IQOQ'^'ii ••'••£;.",
500 ;:'*v~
500
300
500
500
500
500
1000
300
500
1000
500
500
500
300
1000
300
1000
300
300
500
Effluent
Concentration
{mg/L)
115
601
264
579
142
169
43.8
4067
96.1
221
349
16.1
12.1
89.6
50.6
5.4
211
44.9
333
385
120
3.5
0.9
82.7
4.9
-me
•-"6.6' ""
52.8
2.6
1.6
8.8
3.3
1.1
8.8
1.2
3.4
31.6
1.4
1.7
2.0
0.8
21.6
0.8
37.8
5.2
1.6
1.2
Destruction
Efficiency
61.7
39.9
47.2
42.2
71.6
66.3
85.4
59.4
68.0
55.8
65.1
V V •
94.6
97.6
82.1
83.1
98.2
57.8
85.0
66.7
61.5
76.1
99.3
99.7
91.7
99.0
§8:7 ;
89.4
99.1
99.7
98.2
99.3
99.8
99.1
99.6
99.3
96.8
99.7
99.7
99.6
99.7
97.8
99.7
96.2
98.3
99.5
99.8
42
-------
TABLE 6.9 PYRIDINE BATCH TEST RESULTS
Experiment
Number
0731-1
0802-4
0807-2
0810-5
0813-2
0806-2
0802-6
0806-1
0810-1
0810-3
0805-1
0805-6
0808-1
0809-3
0731-3
0731-4
0731-2
0802-3
0807-3
0810-6
0801-4
0806-3
0810-2
0802-5
0808-4
0808-2
0809-2
0801-3
0802-1
0802-2
0807-4
0813-1
0810-7
0801-1
0806-4
0810-4
081 0-A
0805-4
0808-3
0809-1
Reaction
Time
(min)
3
5
5
5
5
5
5
5
5
5
5
5
5
5
6
9
9
10
10
10
10
10
10
10
10
10
10
15
20
20
20
20
20
20
20
20
20
20
20
20
Test
Temperature
ra
404
400
400
406
406
440
450
450
450
450
480
500
500
500
402
401
404
400
400
406
450
450
. 450
480
500 . '
500
500
450 '*"'
400
400
400
400
407
450
450
450
450
500
500
500
Influent
Concentration
(mg/L)
1000
1000
500
750
750
500
1000
500
750
750
1000
1000
500
750
1000
1000
1000
1000
500
750
1000
500
750
1000
1000
500
,,750
1000
1000
1000
500
500
750
1000
500
750
750
1000
500
750
Effluent
Concentration
frng/L)
668
756
353
403
559
56.1
76.5
14.2
82.5
52.4
48.2
14.7
11.3
29.1
648
426
348
356
194
414
67.6
5.1
3,7
11.2
12.0F ":
815 . ":
- **•» . - ,i ' £•
*ffi$'°- :/f-
390
357
251
249
270
11.2
4.3
3.2
1.2
5.3
2.4
6.8
Destruction
Efficiency
(%)
33.2
24.4
29.5
46.3
25.5
88.8
92.3
97.2
89.0
93.0
95.2
98.5
97 7
w I • 1
96.1
35.2
57.4
65.2
64.4
61.3
44.9
93.2
99.0
-99.5
98.9
4-^S:bijro '',.-.'
SS^S&b'i* '^-; •
iXrf-gVw r-;--> j:
61 !o
64.3
1 49.8
50.1
64.0
98.9
; QQ 1
09* t
99.6
99^8
99.5
99.5
99.1
ODS column and a UV detector, set at 229 nm. The mobile phase was acetonitrile-phosphate buffer
solution (40 parts acetonitrile: 60 parts phosphate, pH 2.8) pumped at a flow rate of 2.0 mL/min. In this
study, a Waters' HPLC system with a Supelco O18 column was used and a UV detector was set at 229
nm. Three combination eluents (methanol/acetonitrile/water, acetonitrile/phosphate/water, and
43
-------
acetonttrile/water), each with different mixing ratios, were tested. The HPLC procedure using an
eluent of 30:70 acetonitrile/water ratio was selected for 2,4-D analysis. However, the effluent samptes
showed interfering peaks in the HPLC chromatograms. :
Reportedly, several 2,4-D esters were analyzed by the GC technique (Lamparski and Nestrick,
1978; Supelco 1991 User's Catalog). In this study, analysis of 2,4-D methyl ester using the GC
technique (Hewlett Packard Gas Chromatograph with Electron Capture Detector (ECD) and a Supelco
SPEl-608 capillary column) was successful. The GC detection limit was as low as 0.02 mg/L for 2,4-D
methyl ester. Therefore, 2,4-D methyl ester was used in the batch screening study. I
A total of 44 2,4-D methyl ester batch tests were performed (Table 6,10). Reaction
temperatures and residence times, respectively, ranged from 400°C to 500°C and from 0.67 min to 25
min,, The feed concentrations used were 43 mg/L, 100 mg/L, and 160 mg/L. These low
concentrations were due to the limited solubility of 2,4-D methyl ester in water.
The destruction efficiencies ranged from about 73.3% to 99.7%. Destruction of 2,4-D methyl
ester proceeded rapidly. However, the gas chromatograms of the samples derived from low-
temperature and short-residence time tests indicated the presence of some by-products. The
concentrations of these by-products were too tow to be detected by GC/MS analyses.
6.3 Continuous-flow test
Bas!^^fll-*fet;»l??ISl'^st resuts» two compounds, 2,4-dlchIorbphenoi (DCP) and pyridine,
were selectedjor further cgntinuous-fipw studfes. The results from these studies are described
below :'^^^^W^V::' : str?tS*f'^: - •^•••i- '••-•• •.• ; :^'w&&'<> -fo- : • .•-v.f *;•>:-
The global kinetic expression for both mode! compounds was based on ideal plug-flow tubular
reactor assumptions. In terms of 2,4-DCP or pyridine conversion, X, the equation has two forms:
X = 1 -[1 -Aexp ( (1 - a) &wz&u3QGiQzti(lQ-3 fora* 1 (6.1)
fora=1 ; (6.2)
where A is the pre-exponential factor, Ea is the activation energy, t is the time, C is the concentration
(subscript MC indicates the model compound), and the superscripts a, b, and c are the reaction orders.
A multi-variable non-linear regression technique was used to obtain optimized values for the Arrenhius
parameters, A, Ea, and the reaction orders (a, b, and c).
44
-------
TABLE 6.10 2,4-DICHLOROPHENOXYACETIC ACID METHYL ESTER
BATCH TEST RESULTS
Experiment
Number
1008-10
1008-11
1008-1
1007-3
1001-13
1002-1
1007-2
1007-5
1008-7
1001-5
1001-8
1002-4
1004-5
1009-2
.1009-6
1001-14
1002-2
1007-1
1008-8
1001-6
1003-3
1001-2
1009--5
1004-2
1009-3
1009-7
1002-3
1008-2
1008-4
1008-5
1008-9
1001-7
1003-1
1003-2
1010-2
1004-3
1009-4
1001-10
1009-1
1001-9
1008-6
1009-8
1010-1
1009-10
Reaction
Time
(min)
0.7
0.9
1
1 .
2
2
2
2
2
2
2
2
2
2
2
4
4
4
4
4
4
4
4
4
4
4
8
8
8
8
8
8
8
8
8
8
8
12
12
16
16
16
20
25
Test
Temperature
CO
374
425
374
400
400
400
400
400
400
450
450
450
500
500
500
400
400
400
400
450
450
460
~~ 486
nsoo
566
500
400
400
400
400
400
450
450
450
450
500
500
400
500
400
400
500
450
520
Influent
Concentration
(mg/L)
43
43
160
160
100
75
160
160
43
100
100
160
160
43
100
100
75
160
43
too
.160
too
-100
160
43
100
75
160
43
43
100
100
160
160
43
160
43
100
43
100
43
100
43
100
Effluent
Concentration
(mg/L)
11.3
3.8
22.7
8.7
7.6
5.6
7.7
9.1
2.9
4.9
5.8
5.2
8.0
1.1
4.7
5.7
2.4
3.1
1.4
2.6
5.8
3.4
4.4
5.0
2.2
1.9
1.8
3.0
1.1
0.5
1.7
1.9
2.4
2.7
0.1
2.9
0.8
0.6
0.7
0.5
0.3
0.9
0.2
0.3
Destruction
Efficiency
73.7
91.2
85.8
94.6
|92.5
92.5
'95.2
94.3
93.3
95.1
94.3
96.8
95.0
97.5
95.3
94.3
96.8
98.0
96.8
97.4
96.4
96.6
•: 95.6 '"'
- ...,-[ ."_•••.•.-. . •-. ,
',- Qft G"'**"" * • "
*95!o
98.1
97.6
98.1 .
97.3
98.8
98.3
98.1
98.5
98.3
99.7
98.2
98.0
99.4
98.3
99.5
99.2
99.1
99.5
99.7
45
-------
6.3,1 2,4-dichlorophenol ;
Ten experimental runs were conducted with 2,4-DCP, generating 56 different data points.
The experimental conditions and results for 2,4-DCP are given in Tables 6.11 and 6.12, respectively.
Conversion (X) for the reactions with and without oxygen was calculated as
X = 1 - ([2,4-DCP]f/[2,4-DCP]0). i
i
A. Hydrolysis of 2.4-dichloropheno| :
The reaction of 2,4-DCP with supercritical water was investigated. The maximum conversion of
10.2% was observed for the reaction of 2,4-DCP with SCW at 518°C and 2.2 sec.
The 2,4-DCP hydrolysis data support a pseudo-first-order reaction model expressed as
follows: ;
Xhyd = 1 -exp[1.72 x 1012exp(^) q (6.3)
i
where the units of the pro-exponential factor, A, and activation energy, Ea< are sec'1 and kJ/mo!e,
respectively. The hydrolytic conversion shown ranged from less than 1% to a maximum of 10.1%.
This relatively narrow range of low conversion values is susceptible to error. The average standard
deviation for the analysis of 2,4-DCP standard solutions was 3.2%, and average difference between
the predicted value and the observed value was only 1.8%. The largest devfettoh between a predicted
value and the observed value was 4.95%. Therefore, all of the predicted values were within 2 standard
deviations of the observed values. • ''
TABLE 6.11 CONTINUOUS-FLOW TEST CONDITIONS FOR
2,4-DICHLOROPHENOL
Experiment
Number
1
2
3
4
5
6
7
8
9
10
Nominal Feed
Concentration
(mo/L)
460
780
790
710
820
690
500
560
520
640
Residence
Time Range
(sec)
3.8 to 4.6
3.9 to 4.6
3.9 to 8.1
2.2 to 2.8
4.0 to 7.6
2.1 to 4.1
2.4 to 4.1
7.1 to 13
7.7 to 10.3
7.7 to 7.8
Temperature
Range
(°0
475 to 521
407 to 520
412 to 515
412 to 515
427 to 506
411 to 519
409 to 486
410 to 498
43010511
475
[02]/[DCPJ
Molar Ratio
0 to 8.6
0 to 5.2
0 to 4.5
0 to 5.5
0
5.5
0 to 7.5
0 to 5.9
0 to 5.9
0 to 5.6
46
-------
TABLE 6.12 CONTINUOUS-FLOW TEST DATA FOR 2,4-DICHLOROPHENOL
Sample
ID No.
200
186
196
175
197
191
207
208
204
206
211
210
315
303
308
304
310
313
322
342
329
337
332
306
340
343
371
377
376
397
407
403
405
414
409
411
412
404
423
427
Temperature
PC)
518
521
490
475
451
521
497
478
452
430
408
407
518
496
478
449
429
411
451
519
499
475
450
427
498
477
- 456
431
411
500
472
451
426
410
512
496
452
429
475
476
Xtotal
0.32
0.34
0.26
0.21
0.19
0.27
0.38
0.27
0.11
0.13
0.13
0.21
0.40
0.25
0.22
0.19
0.14
0.13
0.08
0.19
0.15
0.12
0.10
0.11
0.29
0.15
0.13
0.14
0.13
0.43
0.25
0.25
0.18
0.13
0.40
0.30
0.22
0.16
0.18
0.22
Xhyd
0.10
0.12
0.04
0.02
0.01
0.11
0.05
0.02
0.01
0.00
0.00
0.00
0.10
0.05
0.02
0.01
0.00
0.00
0.01
0.06
0.03
0.01
0.00
0.00
0.03
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
Xo,
0.22
0.23
0.22
0.19
0.18
0.16
0.33
0.25
0.10
0.13
0.13
0.21
0.29
0.20
0.20
0.18
0.14
0.13
0.08
0.13
0.12
0.11
0.09
0.11
0.26
0/14
0.12
0.12
0.12
0.42
0.24
0.24
0.17
0.12
0.39
0.29
0.20
0.15
0.16
0.21
Residence
Time
(sec)
3.95
3.98
4.33
4.6
5.23
3.88
4.18
4.5
4.93
6.03
8.1
8.35
4
4.24
4.55
5.25
6.43
7.58
5.18
2.15
2.27
2.51
2.84
3.38
2.29
2.49
2.75
3.25
4.13
, 7.07
7.88
8.78
10.61
12.98
6.8
7.17
8.74
10.3
7.79
7.75
Influent
Cone.
(mg/L)
460
460
460
460
460
780
780
780
780
780
780
780
780
780
780
780
780
780
790
710
790
820
820
790
690
690
500
500
500
640
600
640
640
640
500
500
510
510
690
690
Effluent
Cone.
(mg/L)
310
300
337
360
370
570
480
570
690
680
680
610
430
530
550
570
610
620
720
580
670
720
740
700
490
590
430^:
430
360
450
480
530
550
300
350
400
430
570
530
Oxygen
Cone.
(mole/L)
0.015
0.007
0.016
0.018
0.019
0.007
0.016
0.017
0.018
0.021
0.028
0.014
! 0.015
0.015
0.017
0.020
0.021
j 0.029
! 0.020
0.014
i 0.015
0.016
0.019
0.022
£* •»•«*•
•-"O.O.tS,
'^'fh'M'r
?^M9JrJ22* - "-
7 0.028
1 6.015
i 0.017
; 0.019
; 0.023
0.025
0.014
0.015
0.019
; 0.022
0.005
0.008
Some reaction of 2,4-DCP with SCW, particularly chlorine abstraction, had been expected.
Sequential removal of chlorine atoms from 1,4-dichlorobenzene in supercritical water has been
reported (Jin et al., 1990). In this work, 2-chlorophenol and 4-chlorophenol were detected in effluent
47
-------
samples derived from 2,4-DCP hydrolysis. Some trace levels of phenol were also detected in some of
the hydrolysis samples. Phenol was not detected in the influent or in any of the oxidation samples,
indicating that the chlorine abstraction may proceed through a reaction with H2O. H2O attack on the
aromatic ring was not expected because of the absence of a saturated carbon (Townsend et al., 1988;
Houser et al., 1991). No hydrolysis reaction for phenol was reported under similar 'test conditions
(Thornton and Savage, 1992). The low conversion of 2,4-DCP measured in this work 'supports thes«
findings.
B. SCWO of 2.4-dichlorophenol
Oxidation experiments were conducted at temperatures between 407°C and 520°C. Reaction
pressure was fixed for all experiments at a nominal value of 27.6 MPa, and the feed flow rate was fixed
at 35< gm/min. Reactor residence times varied from 2.1 sec to 10.3 sec. Reynolds number for these
experiments ranged from 7400 to 8200.
I
The overall conversion includes conversions by both oxidation and hydrolysis reactions. The
hydrolytic conversion was calculated using equation (6.3). The simultaneous oxidative Conversion was
calculated
The following kinetic model was derived from regression analysis results with 95% confidence
limits:
r = 10*a*g-0 expC71'^3-7) [2.4-DCPl1*0;25 {Q2jO.38d0.52 ;g: (6.4)r i
where the units of the pre-exponential factor, A, and the activation energy, Ea, are sec"!1
and kJ/mole, respectively. i
The ratio of oxidative conversion to hydrolytic conversion is approximately 3:1 at temperatures
above 500°C. This ratio increases as reaction temperature decreases. At 450°G, the ratio Us
approximately 20:1 . At temperatures below 450°C there was no statistically significant hydrolysis.
C. Overall reaction of 2.4-DCP in oxygenated SOW '
For the development of the overall reaction rate equation, the H2O reaction order was taken as
zero. The rate equation resulting from the regression with 95% confidence intervals is shown below.
r = 105-S±1.8 expC8^25) [2,4-DCP]1±0-25 [O2]0-35±0.36 | (6 5)
The pre-exponential factor for 2,4-DCP is somewhat higher than the values for other phenols reported
in the literature. The pre-exponential factor reported for the SCWO of phenol and 2-chlbrophenol are
48
-------
303 M-1-2«s-1 atxl 102t1-2 M-0-63*s-1, respectively (Thornton and Savage, 1992; Li et al.,
These same authors reported the activation energy for these SCWO processes as 51.0 kJ/mote and
46.0 kJ/mole for phenol and 2-chiorophenol. respectively. The higher activation energy for 2,4-DCP
may be the result of the reduction in the electron density over the aromatic ring. Results from wet air
oxidation studies show that chlorine-substituted phenols have slower oxidation rates (Joglekar ei al.,
1991).
Comparison of model predictions with experimentally observed conversions is shown in
Figure 6.1. The model (equation 6.5) showed a tendency to underpredict the experimentally
observed conversion. The larger than expected conversion of 2,4-DCP may have been the result of a
precipitate that formed in some of the effluent samples within 2 to 24 hrs after collection. The
"8
5
§
iCL
U. JD -
0.3 -
0.25 •
0.2 -
0.15 •
0.1 -
0.05
n
; •
* „
' * • * • *
« • * * - •
•• • .
• ' '"'%.. •.•-.:'•; " •'•
f
• • •
: . -. • .
1 1 11 1 1 1 1
0.05 0.1 0.15 0.2
Observed
0.25
0.3
0.35
Figure 6.1 Comparison of 2,4-DCP Oxidation Conversion:
Predicted vs. Observed.
49
-------
precipitate was extracted from the aqueous phase with methylene chloride, and the organic phase was
analyzed using the HP 5890 GC. The precipitate contained substantial amounts of 2,4-DCP. Effluent
t
samples were analyzed before and after precipitate formation. Some of these samples showed a
decrease in 2,4-DCP concentration of as much as 80% after formation of a precipitate. To minimize the
impact of precipitate formation, all 2,4-DCP effluent samples were analyzed within 6 hrs of collection.
Additional error may have been incurred by difficulties in analysis of the effluent solution.
Many of the transition products generated during the SCWO processing of 2,4-DCP were isomers.
Resolution of the resulting gas chromatograph into distinct and separate species was not possible for
all experimental conditions. Isomers with overlapping signal peaks may have been integrated as a
single species.
Several unsuccessful attempts to generate the precipitate in influent samples were made*.
These attempts included (1) the addition of FeCI3 and HC1 to lower the pH and raise the ionic strength,
(2) lowering the temperature of the effluent influent sample to 4°C, and (3) the introduction of small
amounts of a precipitated effluent. :
D. Transition products from the SCWO processing of 2.4-DCP
Nine different transition products, listed in Table 6.13, were identified in the effluent samples
from the SCWO processing of 2,4-DCP. The first seven compounds were isolated fn most of the liquid
effluent samples. The substituted phenols were identified using a GC/MS (Hewlett Packard system
5895). Identification of the carboxyjic acids was determined by spiking the effluent sample with known
>- "-'k^*'- '*• -..-.** f- • "*"* - .-" "- " , • •••"•., -- ',.--^ --i' "•"^"l >H-1:.-'^''.""
carboxylic acids using the 1C technique. CO and CO2 were isolated in the gaseous effluent. 'The
identity of the gaseous compounds was accomplished by matching residence time with a standard
gas. ',
6.3.2 Pyridine
Eleven experimental runs were made with pyridine, generating a total of 62 data points. The
test conditions and results of pyridine experiments are given in Tables 6.14 and 6.15, respectively.
The [kinetic and mechanistic aspects of hydrolysis and oxidation of pyridine in supercritical water were
derived from these relationships. !
The influent pyridine concentration was the average of at least two samples collected during
each run. Oxygen concentration, as calculated from the mass flow meter readings, represented the
concentration of oxygen at the reactor inlet. The deviation of multiple influent ^concentration
measurements was about 1% for all experiments. Most analyses were made within two days after the
samples were generated. ;
50
-------
A. Hydrolysis of pyridine j
The extent of pyridine hydrolysis in SOW was investigated. The largest hydrolysis impact
corresponded to a 4.2% of pyridine conversion which occurred at 521 °C in less than 7 sec. No
transition products were detected by either GC or 1C analyses. Hydrolysis of pyridine at temperatures
below 500°C was statistically insignificant. These results represented the combined preheater and
reactor effects. However, the hydrolysis effect due to the preheater was insignificant because of the
low temperature profile. •.
TABLE 6.13 2,4-DCP OXIDATION TRANSITION PRODUCTS
Compound
2-chlorophenol
4-chlorophenol
2,6-dichlorophenoi
Phenol
Chloride
Formic Acid
Acetic Acid
Carbon Monoxide
Cartion Dioxide
Structure
C6H4OHCI
C6H4OHCI
C6H3OHCI2
C6H5OH
cr
HCOOH
CH3COOH
CO
• co2
Identification
Method
GC Retention Time
GC Retention Time
GC Retention Time
GC Retention Time
1C Sample Spike
1C Sample Spike
1C Sample Spike
GC Retention Time
GC Retention Time
Retention
Time
(miro
2l7
5L2
5i4
215
3:4
7J4
5JO
30
3.5
•••'..- •,-..- --.- - • -^-v *-' -. '••-.••
TABLE 6.14 CONTINUOUS-FLOW TEST CONDITIONS FOR PYRIDINE
E>periment
Number
1
2
3
4
5
6
7
8
9
10
11
Nominal Feed
Concentration
(mo/U
2007
3050
1134
1990
1990
1261
1819
2875
2215
. 3526
1088
Residence
Time Range
feec>
7.0 - 10.7
6.6 - 10.5
7.1 - 10.5
7.75
6.6 - 8.7
2.2 - 2.8
2.1-2.5
2.2 - 2.8
3.9 - 4.6
3.8 - 4.6
3.8 - 4.6
Temperature
Range
448-516
427-521
426 - 521
475
475 - 522
450-517
450-520
449-540
452 - 526
450 - 525
474 - 522
[02]/[pyr]
Molar Ratio
0
0-1.4
1.6
0.7 - 1.6
1.8
0.9 - 2.1
2.64
0.4 - 2.5
0.9 - i.O
0.5 - 0.9
1.0-2.0
51
-------
TABLE 6.15 CONTINUOUS-FLOW TEST DATA FOR PYRIDINE
Sample
ID No.
708
671
711
606
701
709
718
703
702
700
713
714
754
716
651
715
719
662
666
655
658
682
684
660
680
674
681
673
668
661
676
678
677
685
509
522
520
514
50)3
513
516
525
526
541
519
524
542
529
502
537
535
Reaction
Temperature
510
509
509
501
504
504
475
450
520
502
477
517
518
502
503
476
450
522
499
475
449
522
526
527
498
500
500
474
525
527
527
474
526
497
476
476
476
476
475
522
499
475
452
522
523
499
474
451
427
521
499
Conversion
X
0.22
0.20
0.12
0.08
0.23
0.26
0.09
0.02
0.43
0.21
0.06
0.44
0.40
0.22
0.24
0.09
0.02
0.67
0.33
0.13
0.00
0.60
0.74
0.81
0.34
0.34
0.37
-------
TABLE 6.15 CONTINUOUS-FLOW TEST DATA FOR PYRIDINE, cont.
Sample
ID Ho.
600
604
602
605
528
534
Reaction
Temperature
(°C)
499
476
451
426
501
501
Conversion
X
0.62
0.24
0.07
0.03
0.01
0.01
Residence
Time
(sec)
7.05
7.75
8.78
10.5
7.05
7.05
Influent
[pyr]
(mg/L)
1130
1130
1130
1130
2010
2010
Effluent
[pyr]
(mo/L)
437
866
1060
1110
1984
1975
: [02]
i
(mole/L)
1 .30E-02
1 .43E-02
1.62E-02
1.95E-02
0
0
± 1.65 expC209- 22-4) [pyr]l.0±o.30[o2]0.20±0.i6
B. Oxidation of pyridine \
Based on ten experimental runs using oxygen, pyridine destruction ranged from 2.3% to
95.2%. Figure 6.2 shows the percent of pyridine remaining during SCWO as a funbtion of time for
three temperatures (450°C, 475°C, and 500°C). Multiple data entries for each temperature involved
feed concentrations ranging from 1000 mg/L to 3000 mg/L and oxygen to pyridine ratios ranging from
1 to 2. In general, the pyridine conversion doubted with every 25°C increase in temperature.
The regression analysis using equations 6.1 and 6.2 established the global Kinetic model for
pyridine, shown in equation 6.6. Confidence intervals of 95% were calculated for the specified
parameters by the regression program. !
(66)
C. Reaction transition products and mechanistic aspect ;
A number of transition products resulting from the oxidation of pyridine in SCW were found in
the liquid and gaseous effluents. All transition products identified using the 1C and GC techniques are
summarized in Tables 6.16 and 6.17, respectively. The composite sample used for these analyses
was collected over a period of 15 min under the test conditions specified for Sample No. 668, Table
6.15. Based on these identified transition products, a network of simplified reaction pathways for
pyridine oxidation in SCW was developed. It was assumed that the first step of pyridine oxidation
followed the general initiation mechanism for SCWO of organic compounds. This step involwsd
hydrogen abstraction followed by hydroxylation (Li et al., 1991). K was further assumed that all ring
opening steps were related to the formation of (2-,3-,or 4-) hydroxypyridine and its resonance isomer.
For further evaluation, these reaction pathways were arranged into three categories: a ring opening
step involving 2-hydroxypyridine (Figures 6.3 and 6.4), a ring opening step involving 4-
hydioxypyridine (Figure 6.5), and a ring opening step involving 3-hydroxypyridine (Figure 6.5).
53
-------
pcoposed fraction pathways included several reactions suggesting (he formation of
melhanol. The presence of carbon monoxide in (he gaseous effluent and methano! in the liquid
effluent supported the proposed reaction pathways. The oxidation of methanol and methane in SOW
produced carbon monoxide as an intermediate product (Rofer and Streit, 1989; Webtey and Tester.
1989). Mechanisms for carbon monoxide oxidation and the water shift reaction have been reported
100©
80-
£
« 60
o
•| 40
20-
§
o o
o
Q 500'C
O 475*C
* 450*C
Q
Q
Q
Time (Sec)
10
Figure 6.2 Percent of pyrldlne remaining during SCWO as a function of time
at 450'C, 475°C, and 500°C (pressure = 27.6 MPa, feed concentration =
1000 • 3000 mg/L, |O2J vs. fpyr] = 1.0 - 2.0).
54
-------
TABLE 6.16 PYRIDINE OXIDATION
Compound ~ ~~
Borate Eluent
Propanoic Acid
L-Lactic Acid
Acetic Acid
Glycolic Acid
Acrylic Acid
Foimic Acid
Carbonate Eluenj
Monocarboxylic Acids
Unknown A
Nitrite
Unknown B
Maleic Acid
Structure
CHaCHaCOOH
CH3CH(OH)COOH
CHaCOOH
HOCH2COOH
CH2 = CHCOOH
HCOOH
(Mixture of the above i
NO2-
HOOCCH = CHCOOH
TRANSITION PRODUCTS IDENTIFIED BY 1C
pKa!
4.87
3.08
4.75
3.83
4.25
3.75
6 compounds)
(cis) 1.83
pKaa Elution Time
(min)
K n
w. w
5 O
\jt\i
; K n
o.u
K fi
•J.U
ft ft
D.O
7.4
2.0
3.4
A C
*T.sJ
4.8
6.07 7.5
(Helling and Tester, 1988; Holgate et al., 1992). Since hydrogen has also been found as a transition
product in SCWO of methanol (Webley and Tester, 1989), it was proposed that some transition
products shown in Figures 6.3 through 6.5 were derived from hydrogen interaction.
Maleic acid was found In most liquid effluent samples characterized by relatively high pyridine
conversion. The highest concentration observed for maleic acid was:about 50 mg/L. The presence of
this acid clearly indicated that at least one nitrogen - carbon bon|was broken in order to preserve a
four-carbon molecular chain. However, the presence of dimethylamine suggested that the"bond
cleavage occurred between carbon atoms. Therefore, it appeared that ring opening and
fragmentation were random occurrences. i
i
Although pyridine hydrolysis was negligible, water may have played an important role in the
pyridine oxidation reactions. Some of the transition products shown in Figures 6.3 to 6.5 could be
reactive with water. Such a reaction would be highly favorable because the number of Water molecules
is large compared to the number of oxygen molecules and active radicals. Therefore, once the
pyridine ring was fragmented, there could have been at least four competing reactions: oxidation,
hydrolysis, hydrogenation, and decarboxylation.
The presence of nitrogen transformation pathways was supported by evidence of various
nitrogen-containing transition products in the reactor effluents. Ammonia (NH3) was a major transition
product in the liquid effluent. Also, dimethylamine in the liquid effluent was confirmed by GC-MS
analyses. The 1C measurements suggested possible existence of trace amount of nitrite (NO2-).
55
-------
TABLE 6.17 PYFHDINE OXIDATION TRANSITION PRODUCTS IDENTIFIED BY GC
Compound
Structure
Molecular Weight Reverse Fit
(g/mole)
Retention Time
(min)
eci
Methanol
Acetone
Unknown A
Dimethylamine
Unknown B
GC-MS**
CH3OH
CH3COCH3
(CH3)2NH
Liquid Effluent
32
58
45
0.63
0.78
0.80
0.86
0.97
Dimethylamine (CHsfoNH 45 841
Formamide HCONH2 45 824
Ethyiamine CH3CH2NH2 45 767
GC-MS***
FoimicAcid HCOOH 46 690
Acetic Acid CHsCOOH 60 730
Gliitaconic Acid HOOCCH2CH = CHCOOH (cis) 130 770
MaleteAcid HOOCCH = CHCOOH (cis) 116 720
SuccinteActd HOOCCH2CH2COOH 118 880
GlutaricAcid HOOCCH2CH2CH2COOH 132 810
MalonicAcid HOOCCH2COOH 104 720
Propionic Acid CH3CH2COOH 76 930
Oxalic Acid HOOCCOOH 90 750
Gaseous Effluent
GS .
Canton Dioxide CO2 44
Cairbon Monoxide CO 28
Methane CH4 1 6
* Samples were analyzed without using solvent extraction.
** Samples were extracted using methylene chloride and then analyzed
*** Samples were ether extracted and methyl esterified with boron trifluoride.
However, no nitrogen products were found in the gaseous effluent using the on-line
These results were in general agreement with reported information. According to p«
studies, nitrogen remained as ammonia or ammonium at relatively low temperature
sut)critical temperature for water to about 450°C (Takahashi and Isobe, 1988; Takahs
Tiffany et al., 1984). At higher temperatures (560 - 670°C), nitrogen (N2) and nitrous o>
dominant nitrogen products (Timberlake et al., 1982; Killilea et al., 1992).
One ozonation study involving an aqueous solution of pyridine revealed a nur
intermediates and by-products (Andreozzi et al., 1991). These compounds included ai
0.72
0.72
0.72
7.6
8.1
28.8
23.8
23.9
29.5
24.4
32.0
26.2
3.5 .
30
17
gas partitioned
jvious oxidation
5S ranging from
shi et al., 1989;
dde (NaO) were
nber of reaction
nmonia, nitrate,
56
-------
Eihyf
Atninc*
3-Hydroxy
Ptopaaoic
Acid
Mgure 6.3 Simplified reaction pathways for pyrldlne oxidation In supercritical
water-ring opening step Involving 2-hydroxypyrldIne Intermediate and
carbon-nitrogen bonds (* compounds Identified In this study).
57
-------
Pyridiuc
2-H ydroxy-
pyridinc
V
°-WH
Glyoxal
oxanxic Add
1 1
H-N
VCH3
Dimethyl
Aminc*
\ £>H
vc
CH2
A
Malonic
Acid
V
O. OH
X
O OH
Oxalic
Acid*
/
NHs
+
O. OH
V
i
O OH
Oxalic
Acid
O OH
C/VNH2
Oxanuc
Acid
/ 1
O
^Cv
h NH2
Formamide*
I
|
«
H^^^^/^i 1
\J\\
Fonnic
Acid*
NHs*
CH3
OH
Meibanot*
.
co»
I
I
OOa
-
•*^^^*t
CH3
0AOH
Acetic
Add*
\ ^
O
II
t •X'VN.y^l
H Ol
Fonnic
Acid*
1
CO2 + H2<
o r»w
T
O* SOH
Oxalic
Acid | -
1 g
i
j
I
Figure 6.4 Simplified reaction pathways for pyrldlne oxidation In supercritical
water—ring opening step Involving 2-hydroxypyridIne Intermediate and
carbon-carbon bonds ('compound identified In this study).
58
-------
[
/
A
^C Cx
V^OH
i
1
H
HO^O
|\
1
NH2
CHa
YH
A
O OH
Oxalic
Acid*
^^T*"
/x.
X
= &-
^XIH
3-Hydroxy-
pyndiiie
X
NH
^^ f
|
1
NH2
CH2
C
JSJ M
j* ^^^ ^ ^X.
-p — (p ^^
OH
P\nVl<5llK
/
"''OH V^X
*O NH3-+ CHa
i
(/%H
O OH Malouic
*(f Acid
1 / \
A_Q / \
i~ \. / x
^
0
o ;
/ \
. \
iH
i
H3C" NCH3
Oiniethyl
A mine*
; +
H3CXc,CH3
u
O
O OH £ \^ Jf Acetone*
Ct OW *-t\J<£ ':
Glycinc v *-*" r\ f\u .4. I
\&L S*^ 1
i \ Y 5* T
COa*
Vctte
A
O H
N-Mcthyf-
fonuamide
{
CO*
.J.
NHa
CHa
\
t ^
COa
NHa
CHa
Methyl .
• °» .
1
I
NH3
"*"
CH3
r
OH
Methanol*
V*S. <£
Ox OH O^
v CH3
OH *
NHa/^ S? J3? M^«"
VH \ / Jia
Ao- ft ,c
\jfiZ \J
|iXVS^»l|
dycoGc . n vn
i Acid*
* 1
COa I
^ . COa + HaO
CHa
Methafiol*
O OH
Acetic
Acid*
1
l
[
Figure 6.5 Simplified reaction pathways for pyridlne oxidation In supercritical
water—ring opening step Involving 4- and 3-hydroxypyrldIne Intermediates
('compounds identified In this study).
59
-------
pyridiine N-oxide, N-formyl-oxamic acid, and oxamic acid. In this study, trace amounts of nitrite instead
of nitrate were found. Oxamic acid was not found in the liquid effluent. Considering their relative
thermal stability, no effort was made to check the presence of pyridine N-oxide and N-formyl-oxamic
acid. :
As shown in Tables 6.16 and 6.17, two peaks in the 1C chromatograms and two peaks in the
GC chromatograms were not matched with a long list of suspected transition products. The peak areas
for these unknown compounds were relatively small. Samples containing higher levels of transition
products generally displayed a light pinkish colon components causing this color were not identified.
6.3.3 Summary of kinetic models
Kinetic models derived from batch and continuous-flow test data for acetic acid, 2,4-DCP, and
pyridine are summarized in Table 6.18. The errors at the 95% confidence level for each parameter are
provided. The variability of these kinetic parameters obtained from this work is comparable with those
reported in the literature. . ;
TABLE 6.18 SUMMARY OF SCWO KINETIC MODELS DERIVED FROM THIS
WORK AND THE LITERATURE
Compounds Reactor
Type
Kinetic Parameters*
=a
n
m
Temp. Pressure [C]o
CC) (MPa)| (g/L)
Models from this work
AcellicAcid batch lo4-9*0-2 101±24 1
2,4-DCP flow 105-5±1-8 89±25 1±0.25
Pyridine flow 1013-1±1-65 210±22 1±0.30
0 400-500
0.35±0.36 400-520
0.20±0.16 430-530
29-55 &5-3.0
27.6 0.3-0.8
27.6 0.2-3.0
Models frpm literature \\ •
Acetamide1 flow i04-7±4-6 95±23 1 0.17±0.26 400-525 23-3410.01-1.1
Ammonia2 flow lO6-5*3-6 157±65 1 0 530-700 24.6 -0.03-0.12
2-CP3 flow 102-041-2 46±16 0.88±0.06 0.41±0.12 300-420 19-300.01-0.24
Mettianol4 flow 1028-8±10-5 447±125 0.89±0.69 0.12±0.66 454-544 24.6 0.04-0.18
Methanol4 flow 1026-2*5-8 408±85 1 0 454-544 24.6 0.04-0.18
* Kinetic parameters are defined by -d[CJ/dt = k[C]n[Olm and k = k'exp(-Ea/RT), where [C] and [OJ
are concentrations of organic reactant and oxidant, respectively; Ea is in kJ/mole; T is in K; R =
8.314J/mole-K;andk* = (mole/L)l-n"m/sec. [C]o = feed concentration.
1. L.ee et al., 1991. Hydrogen peroxide was the oxidant.
2. VVebleyetal., 1991.
3. Li et al., 1993. The model also includes a reaction order of 0.34±0.17 in water.
4. 2-chlorophenol (2-CP), Tester et al., 1993. !
60
-------
6.4 Corrosion tests
Extremes in pH, high concentrations of dissolved oxygen, ionic inorganic species, and high
temperature-pressure conditions enhanced corrosion. The corrosion by-products In the effluent
caused two potential problems. First, corrosion products such as chromium may affect the quality of
the effluent and the ash. Second, a variety of corrosion products, usually metal oxides, increased the
plugging potential of pressure-regulating devices and the potential for erosion. ;
The material evaluation study was designed to investigate potential chloride problems. Three
i
alloys (SS 316, Hastelloy C-276, and Monel 400) were surveyed using a chloride concentration of 420
mg/L and exposure times of 100 hrs. Table 6.19 summarizes uniform corrosion rates for these alloys
in subcritical and supercritical water conditions. Generally, the highest corrosion rates occurred at the
subcriticaltemperature, 300°C, and the lowest pH condition, pH = 2.1. The higher corrosion rates at
subciritical conditions were explained by the large amount of dissociated ions in solution and possibly a
higher dielectric constant. The corrosion rates increased, regardless of pH condition, at higher
temperatures. Higher corrosion rates, at 500°C, reinforced the temperature dependence of the
electrochemical and chemical reactions. As the temperature increased, the. rates of the chemical
reactions involved in the corrosion process also increased. At a subcritical temperature of 300°C, the
properties of water appeared to facilitate electrochemical corrosion. As suggested by the higher
dielectric constant, inorganic ions were free in solution because of the existence of the highly polar
wafer molecules. However, at temperatures above 400°C, the dielectric constant of water and the
inorganic solubility dropped off markedly, Therefore, it was concluded that chemical corrosion was
dominant at SOW conditions because neither HCI, NaCI, nor NaOH waa highly dissociated.
Temperature affected the average uniform corrosion rate. Stainless Steel 310 and Hastelloy
C-276 experienced higher corrosion rates at 300°C (subcritical) and 500°C as compared to 400°C,
regardless of pH level. At subcriticat conditions, the inorganic species such as NaCI and HCI were
dissociated and the oxidizing and halide ions were free in solution. The dielectric properties of the
water facilitated electrochemical corrosion. .At higher temperatures, chemical reactions involving
corrosion proceeded at a faster rate. .
Crevice corrosion can be minimized by designing a reactor system such that fluid does not
stagnate. This includes using butt welds for all joints to assure that stagnant pockets! of fluid cannot
develop. Conversely, pitting cannot be prevented by design. Some type of chemical or electrical
inhibition might eliminate or reduce this type of corrosion. Also, metal coatings or tube liners may be
used to reduce pitting.
Higher average uniform corrosion rates and more extensive localized corrosion were
experienced by the coupons exposed to the low pH solution. Generally, the three metals examined in
61
-------
TABLE 6.19 UNIFORM CORROSION RATES FOR THREE ALLOYS IN SUBCRITICAL
AND SUPERCRITICAL WATER
Alloy Temperature
fC)
SS 316 300
400
500
Hastelloy C-276 300
400
500
Mohel400, 300
400
, .
500
Initial Test Solution Corrosion Rate !
PH
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
2.1
5.8
8.6
(mm/yr)
0.0479
0.0415
0.0149
0.0209
0.00085
0.00596
0.0219
0.00915
0.00703
0.0162
0.0117
0.0121
0.0119
0.00149
-0.00192
0.0285
0.0339
0.00298
0.0818
0.0281
0.0316
0.675
0.481
0.449
-0.291
-o.aie
-0.174
(mpy)
1.89
1.63
0.59
o;82
OI03
0:23
0:86
0.36
0:28
0:64
0^46
0.48
0:47
Oi06
-Oi08
1,12
1;33
0.12
3J22
26.59
-11148
im
8.94
-12.43
5ll8
17,67
-6;86
Notes: Chloride concentration in feed solution = 420 mg/L; fluid density = 0.09 to 0.3 g/ml; exposure time = 100
hours. ' ' •
...... ..... .|.
this study experienced higher corrosion rates, regardless of temperature, in the experiments with the
lowest pH. Upstream neutralization could be used to eliminate extremes in pH and reduce metal loss.
Spectrographic analysis of the surface of the Monel 400 coupons, after exposure to the test
solutions at 500°C, revealed a significant increase in copper concentrations and a decrease in nicked
concentrations. The Monel 400 coupons actually increased in weight. This deposition occurred
because copper leached from the reactor walls. The Monel 400 reactors were a greater source of
metal than the coupons and possibly had been sensitized during the previous low-temperature tests.
Monel 400 samples exposed to the test solutions at 400°C showed unacceptable average uniform
corrosion rates. !
62
-------
6.5 Chromium speciation and separation
The identification of chromium corrosion species generated during SCWO was a major effort in
this study. Experiments were conducted using a bench-scale, vertical, concentric-tube;reactor system
and municipal wastewater sludges. The reactor material was SS 316 (18 wt% chromium). The
objective was to determine the hexavalent and soluble trivalent chromium concentrations in the
treated effluent and in sediments in the reactor bottom.
As the SCWO tests progressed, hexavalent and trivalent chromium corrosion products were
generated. Both chromium species were removed by precipitation. At 400°C and a Reynolds number
of approximately 8000, the hexavalent and soluble trivalent chromium concentrations in these
effluents were <0.004 mg/L and 0.163 mg/L, respectively. After 8.76 kg of sludge was processed,
the concentrations of hexavalent and soluble trivalent chromium in the bottom of thb reactor were
0.2813 mg/L and 3.72 mg/L, respectively. The results for municipal and industrial sludges, including
test conditions, chromium concentrations, COD concentrations, and pH, are summarized in Tables
6.20 and 6.21, respectively. Literature data involving pH-potential (Pourbaix) diagrams of Cr-HaO-O2
systems at standard and critical conditions for water suggest that there may be a correlation between
chromium species concentration and the pH of the influent and the effluent.
Chromium separation efficiency, species distribution, and corrosion product depended on
sludge characteristics and treatment conditions. Two mechanisms were responsible for separation of
hexavalent and trivalent chromium from effluents: (1) the precipitation of chromate complexes and
trivalent chromium salts and (2) the settling of solid residues onto which trivalent ehrpmhtrn hart
adsorbed.- . •"-'.• : - • -,. :.; :;."': '"It^T
In the case of municipal sludge, hexavalent chromium, In the form of insoluble chromate
complexes, settled from the bulk supercritical fluid. The precipitate simply dropped out from the
effluent as a result of near laminar-flow conditions. While the concentration of hexavalent chromium iirt
the effluent was <0.004 mg/L, the concentration of hexavalent chromium in the reactor bottom was ait
least 72 times greater. At subcritical and laminar-flow conditions, concentrations of chromate
complexes in the effluent and reactor bottom were 0.046 mg/L and 0.035 mg/L, respectively.
For both temperature regimes used during treatment of the industrial sludge, the
concentration of hexavalent chromium was below the detection limit in both the effluents and reactor
bottom. A large portion of trivalent chromium was sorbed onto solid residues collected in the reactor
bottoms. Concentrations of soluble trivalent chromium in the effluent and reactor bottom were
comparable. !
63
-------
TABLE 6.20 CHROMIUM CONCENTRATIONS DERIVED FROM SUB-CWO AND
SCWO OF ANAEROBICALLY DIGESTED MUNICIPAL SLUDGE
Sampling Reaction Zone Chromium Concentration
Location Temperature Cr46
(°C) (mo/U
influent
Effluent
Reactor Bottom
Influent
Effluent
Reactor Bottom
300
300
300
400
400
400
<0.004
0.046
0.035
<0.004
<0.004
0.288
Soluble Cr+3
(mg/L)
<0.01
0.390
0.710
<0.01
0.607
3.712
Total
Insoluble Cr*3 COD
(mg/L) (ma/U
0.402
0.319
0.618
0.541
0.163
2.130
10300
2270
972
11300
2640
1170
PH
7.3
7.9
7.8
7.4
7.9
7.7
Notes: (1) Feed flow rate = 120 g/min; pressure = 24.8 MPa; (2) Reaction zone temperature was
nearly isothermal (within ± 10*C).
TABLE 6.21 CHROMIUM CONCENTRATIONS DERIVED FROM SUB-CWO AND
SCWO OF INDUSTRIAL SLUDGE
Sampling Reaction Zone Flow
Chromium Concentrafirm
Location Temperature Rate Cr46 Soluble Cr*3 Insoluble Cr*3
CQ (q/min) (mo/L) (mo/U (ma/L)
Influent
Effluent
Reactor Bottom
Influent
Effluent
Influent
Effluent
Reactor Bottom
367
367
367
400
400
450
450
450
_
90
90
—
95
— .
95
95
<0.004
<0.004
<0.004
<0.004
<0.004
<0.004
<0.004
<0.004
-------
Pressure gradients through a letdown valve can cause captation and subsequent valve failure.
These gradients can also cause physical erosion by propelling particles at high velocities towards valve
surfaces. For a given pressure drop across letdown valves, the fluid velocity dependsion the process
throughput, fluid density, and valve characteristic geometry. In most applications, temperature at the
pressure letdown valve will be well below supercritical water conditions. The process economics
based on energy calculations dictates that as much heat as possible be transferred from the reactor
effluent to the feed stream. The average temperature at the letdown valve would be approximately
50°Cto80°C. i
The SCWO process effluent may have a high solids content. The nature of the solids
depends on the feed stream, but could include metal oxides, inorganic salts, and/or residual organic
matter. These particles can cause erosive wear and plugging in letdown valves. Research has shown
that micron-sized particles can be effectively removed under SCWO conditions by k cyclone-type
solids separator (Dell'Orco and Gloyna, 1992). Other solid-fluid separation devices may also be used;
therefore, solids handling features of a letdown valve must be flexible and adaptable.
There may be significant amounts of dissolved gases (such as Na, ©2, CO2. and NHs) in the
SCWO effluent Gases can not only propel particles at high velocities, but also create a phenomenon
called vapor cavftation. Both actions can cause erosion. j
The development of pressure letdown devices for SCWO applications should focus on
improved valve designs and selection of proper materials. Operating a laboratory-scale pressure
letdown device is more challenging than operating a pilot-scale or commercial-scale: one Because
smaller valves have less flow and passageway to regulate. Therefore^ ^^g sr^^b^^^rri^piij
a scale as large as possible. " : ' ' '' - .;•""' '"•"?•-'' ^-^^-^(i^^^^^
- -
6.7 Process development -
A generic SCWO process involves an aqueous feed stream containing liquid organic
substances or a pumpable organic sludge, an oxldant source, and possibly supplemental fuel or other
additives. Trie oxidant may be air, oxygen, hydrogen peroxide, or perhaps a chemical compound
containing oxygen. In cases where the heating value of the waste is insufficient, supplemental fuel
may be substituted for an external heat source. By using properly designed heat exchangers, and a
waste containing 2% to 5% organic compounds, the heat developed by the oxidation can maintain the
required reaction temperature. Supplemental fuel may be inexpensive organic materials such as
biological sludges, waste oils, or other organic compounds. Conversely, if the wastes are too
concentrated the influent may be diluted using the effluent water.
65
-------
The key to a successful SCWO process design is integration of various unit operations. Basic
design aspects are as follows:
• process configuration and reactor type; I
• reactor residence times, pressures, and temperatures;
• materials of construction for each unit operation;
• control and removal of solids from the treated effluent; and
• operations and maintenance of the facility, including safety, analytical support, and regulatory
monitoring/disposal requirements.
It is also important to consider characteristics of a particular feed stream in the overall procejss
design. For example, concentrations of inorganic salts, metal oxides, and chlorides in a feed stream
may dictate reactor design, material selection, and solids separation techniques required.
Kinetic and mechanistic information generated in this work can be used In assisting more
efficient reactor design and selecting optimum process conditions. By adjusting the reaction
conditions, the type and amount of transition compounds can be controlled, hence achieving higher
destruction efficiency.
66
-------
1
7. Information Transfer
Since the initiation of this project in 1990, much progress in SCWO technology development
has been made. Through rigorous and extensive R&D effort, UT researchers have pushed SCWO
technology from laboratory-scale operation to the edge of commercialization. These accomplishments
have contributed to an effective information/technology transfer program.
Much information has been channelled to the industrial community through the Separations
Research Program (SRP), The University of Texas at Austin. SRP is an academic/industrial consortium
supported by 30 corporate sponsors. One of the SRP sponsors, Eco Waste Technologies (EWT),
Austin, Texas, began joint R&D projects with UT in 1990. EWT provided funding to design, construct,
test, and operate an SCWO pilot plant. This 40 gph SCWO facility provided an opportunity for UT and
EWT researchers to test and evaluate various process components and to conduct, realistic industrial
treatability studies. The integration of complex unit processes and control systems made ft possible to
evaluate pilot plant performance under simulated field conditions. Based upon this general design
and operating experience, EWT has now designed the first commercial SCWO facility. Full-scale
operation is projected to commence in 1994....--.. ;
Also, technology transfer was supported by an active information exchange and technology
demonstration. Visitors from a number of Federal arid State Agencfes/private companies; and foreign
institutions visited UTs SCWO facility. One of tfiesev visitors was the Assistant Secretary of the Navy.
During the course of this project, five master's theses were completed and two national
presentations were delivered. In addition, interim findings were reported at four SRP conferences. A
list of these documents is provided in Appendix A.
67
-------
REFERENCES I
Andreozzi, R., Insola, A., Caprio, V., and D' Amore, M.G. Ozonation of pyridine in aqueous solution:
mechanistic and kinetic aspects. Water Research. 26(6): 655,1991. '
Armellini, F.J. and Tester, J.W. Solubilities of sodium chloride and sodium sulfate in sub- and
supercritical water vapor. Paper presented at the American Institute of Chemical Engineers
National Meeting, Los Angeles, California, November 17-22,1991.
Bird, R.B., Stewart, W.E., and Lightfoot, E.N. Transport phenomena. Wiley: New York; 1960.
Bischoff, J.L., Rosenbauer, R.J., and Pteer, K.S. The system NaCI-HaO: Relations of vapor-liquid
near the critical temperature of water and of vapor-liquid-halite from 300- 500°C. Geochim
Cosmochim. Acta. 50:1437,1986.
Bramlette, T.T., Mills, B.E., Hencken, K.R., Bryniidson, M.E., Johnston, S.C., Hruby, J.M., Feemster,
H.C., Odegard, B.C., and Modell, M. Destruction of DOE/DP surrogate wastes with supercritical
water oxidation technology. SAND90-8229. Sandia National Laboratories, November 1990.
Clesceri, L.S., Greenberg, A.E., Trussell, R.R., and Franson, M.A.H. Standard methods for the
examination of water and wastewater. 17th ed., American Public Health Association.
Washington, D.C., 1989.
Connolly, J.F. Solubility of hydrocarbons in water near the critical solution temperature. Journal of
Chemical Engineers Data. 11(3): 13,1966.
Dell'Orco, P., and Gloyna, E.F. The separation of particles from supercritical water oxidation effluents.
Center for Research in Water Resources, The University of Texas at Austin, Technical Report
235, 142 p., 1992.
Freeman, H.M. Innovative thermal hazardous organic waste treatment processes. Noyes: Park Ridge,
New Jersey, ChapV4,1985.
Gloyna, E.F. and Li, L Supercritical water oxidation: an engineering update. Paper presented at the
EPRI-NSF Symposium, San Francisco, California. February 22-24,1993.
Gloyna, E.F., Li, L., and Bravo, J.L. Destruction of aqueous hazardous wastes in supercritical water.
Paper presented at the 2nd International Symposium on High Pressure Chemical Engineering,
Erlangen, Germany, September 1990.
Helling, R.K. Oxidation kinetics of simple compounds in supercritical water: carbon monoxide,
ammonia, and ethanol. Sc.D. thesis. M.I.T., Cambridge, Massachusetts, 1986.
Helling, R.K., and Tester, J.W. Oxidation kinetics of carbon monoxide in supercritical water. Journal of
Energy and Fuels. 1(5): 417,1987.
Helling, R.K., and Tester, J.W. Oxidation of simple compounds and mixtures in supercritical water:
carbon monoxide, ammonia, and ethanol. Environmental Science and Technology. 22(11):
1319, 1988.
Holgate, H.R., Webley, P.A., Tester, J.W., and Helling, R.K. Carbon monoxide oxidation in
supercritical water the effects of heat transfer and the water-gas shift reaction on observed
kinetics. Energy and Fuels. 6: 586,1992.
68
-------
Houser, T.J., Zhou, Y., Tsao, C.-C., and Liu, X, The removal of heteroatoms from organic compounds
by supercritical water. Paper presented at American Institute of Chemical Engineers Annual
Meeting, Los Angeles, California, November, 17-22,1991.
Japas, M.L., and Franck, E.U. High pressure phase equilibria and PW-data of the water-oxygen
system including water-air to 673 K and 250 MPa. Ber. Bunsenges. Phys. Chem. 89: 12(58
1985.
Jin, IL, Shan, Y., and Abraham, M. The effect of supercritical water on the catalytic oxidation of 1,4-
dichlorobenzene. Journal of Supercritical Fluids. 3: 233,1990.
Joglekar, H.S., Samant, S.D., and Joshi, J.B. Kinetics of wet air oxidation of phenol ;and substituted
phenols. Water Research. 25(2): 135,1991. :
Killilea, W.R., Swallow, K.C., and Hong, G.T. The fate of nitrogen in supercritical water oxidation.
Journal of Supercritical Fluids. 5(1): 72,1992.
Lamparski, L.L. and Nestrick, TJ, Determination of trace phenols in water by gas chromato-graphic
analysis of heptafluorobutryl derivatives. Journal of Chromatography. 156:143,1978.
Lee, D.S. Supercritical water oxidation of acetamide and acetic acid. Ph.D. dissertation. The
University of Texas at Austin, August 1990. ;
Lee, D.S., Kanthasamy, A., and Gloyna, E.F. Supercritical water oxidation of hazardous organic
compounds. Paper presented at the American Institute of Chemical Engineers Annual Meeting,
Los Angeles, California, November 17-22,1991.
Li, L., Chen, P., and Gloyna, E.F. Generalized kinetic model for wet oxidation of organic compounds.
American Institute of Chemical Engineers Journal. 37(11): 1687,1991.
Li, l... Eller, J., and Gloyna, E.F. Pilot-scale reactor systems for supercritical water oxidation of
wastewater and sludge. Paper presented at the American Institute of Chemical Engineers
Summer National Meeting, San Dfega, California, August 19-21,1990. :
LI, Ft, Savage, P. E., Szmukfer, D. 2-Chtorophenof oxidation In supercritical water, global kinetics and
reaction products. American Institute of Chemical Engineers Journal. 39(1): 178,1993.
Marlynova, O.I. Solubility of inorganic compounds in subcritical and supercritical water. In D. de G.
Jones, J. Slater, and R.W. Staehte (eds.), High Temperature High Pressure Electrochemistry in
Aqueous Solutions. National Association of Corrosion Engineers Conference, NACE-4, 131.
University of Surrey, England, January 1976.
Modell, M. Treatment of pulp mill sludges by supercritical water oxidation, final report CE/40914-T1.
Department of Energy Report, July 1990.
Modell, M., Gaudet, G.G, Simson, M., Hong, G.T., and Biemann, K. Supercritical water testing reveals
new process holds promise. Solid Wastes Management. 25:26,1982.
Oh, K.-H., and Tuovinen, O.H. Degradation of 2,4-dicholorophenoxyacetic acid by mixed cultures of
bacteria. Journal of Industrial Microbiology. 6: 275,1990.
Refer, C.K., and Stre'rt, G.E. Oxidation of hydrocarbons and oxyenates in supercritical water. Phase II
Final Report. LA-11700-MS, DOE/HWP-90. Los Alamos National Laboratory, 1989.
Shanableh, A.M. Subcritical and supercritical water oxidation of industrial, excess activated sludge.
Ph. D. dissertation. The University of Texas at Austin, 1990. ;
69
-------
Shaw, R.W., Brill, T.B., Clifford, A.A., Eckert, C.A., and Franck, E.U. Supercritical water: a medium for
chemistry. Chemical & Engineering News. Special report, December 23,1991.;
Smith, J.M., and Raptis, TJ. Supercritical deep well wet oxidation of liquid organic wastes Paper
presented at the International Symposium on Subsurface Injection of Liquid Wastes, March
1986.
Takahashi, Y. and Isobe, S. The catalytic wet-oxidation of ammonium acetate for CELSS Paper
presented at the 16th International Symposium on Space Technology and Science, Sapporo,
1988.
Takaihashi, Y., Wydeven. T., and Koo, C. Subcritical and supercritical water oxidation of CELSS model
wastes. NASA Conference Publication 10040: 95-106,1989.
Tester, J.W., Holgate, H.R., Armellini, F.J., Webley, P.A., Killilea, W.R., Hong, G.T., and Barner H F
Supercntical water oxidation technology: a review of process development and fundamental
research. In D.W. Tedder and F.G. Pohland (eds.), Emerging Technologies; for Hazardous
Waste Management III, ACS Symp. Ser. 518, Chapter 3. American Chemical Society
Washington, DC, 1992. : '"
Tester, J.W., Webley, P. A., Holgate, H.R. Revised global kinetic measurements of methanol oxidation
in supercntical water. Industrial and Engineering Chemistry Research. 32(1): 236,1993.
Thiell, R., Dietz, K.H., Kerres, H., Rosenbaum, H.J., and Steiner, S; Process for wet oxidation of
organic substances. U.S. Patent No. 4,141,829, February 1979.
Thomason, T.B., and Model!, M. Supercritical water destruction of aqueous wastes. Hazardous
Waste. 1(4): 453,1984.
Thornton, T.D., and Savage, P.E. Phenol oxidation in supercritical water. Journal of Supercritical
Thornton, T.D., and Savage, P.E. Kinetics of phenol oxidation in supercritical water. American Institute
of Chemical Engineers Journal. 38(3): 321,1992. . ;.
Tiffany, D.M., Houser, T.J., McCarville, M.E., and Houghton, M.E. Reactivity of some nitrogem-
F° ICh"9 ?°mpound!? at supercritical water conditions. American Chemical Society, Division cif
Timbertake, S.H., Hong, G.T., Simson, M., and Modell, M. Supercritical water oxidation for wastewater
treatment: Preliminary study of urea destruction. Society of Automotive Engineers Technical
Paper Senes No» 82872,1982. • >
Tongdhamachart, C. Supercritical water oxidation of anaerobically digested municipal sludqe Ph D
dissertation. The University of Texas at Austin, December 1990.
Townsend, S., Abraham, M., Huppert, G., Klein, M., and Paspek, S. Solvent effects during reactions
in supercritical water. Industrial and Engineering Chemistry Research. 27:143,1088°
Webley, P.A., and Tester, J.W. Fundamental kinetics of methanol oxidation in supercritical water. In
K.P. Johnston, J.M.L. Penninger (eds.), Supercritical Ruid Science and Technology American
Chemical Society Symposium Series 406, Chap. 17. American Chemical Society. Washington,
DC, 1989.
Webley, PA, Tester, J.W., and Holgate, H.R. Oxidation kinetics of ammonia and ammonia-methanol
mixtures in supercntical water in the temperature range 530-700°C at 246 bar. Industrial and
Engineering Chemistry Research. 30(8): 1745, 1991. i
70
-------
Wightman, T.J. Studies in supercritical wet air oxidation, M.S. thesis. The University of California,
Berkeley, California, March 1981. j
Wilmanns, E.G., Li, L, and Gloyna, E.F. Supercritical water oxidation of volatile!acids. Paper
presented at the American Institute of Chemical Engineers Summer Meeting, Philadelphia,
Pennsylvania, 1989.
Yang, H.H., and Eckert, C.A. Homogeneous catalysis in the oxidation of p-clilorophenol in
supercritical water. Industrial and Engineering Chemistry Research. 27: 2009,1988.
71
-------
APPENDICES
I
A. List of Publications and Presentations
"Corrosion Behavior of Three High-Grade Alloys in Supercritical Water Oxidation Environments" Chris
Matthews, MS Thesis, Department of Civil Engineering, The University of Texas at Austin
August 1991. '
"Erosion Control in Supercritical Water Systems," David A. Sheets, MS Thesis, Department of Civil
Engineering, The University of Texas at Austin, August 1991.
"Supercritical Water Oxidation of Hazardous Organic Compounds,' Dong Soo Lee Anasuya
Kanthasamy, and Earnest F. Gloyna, presented at the AtChE Annual Meeting, Los Angeles
CA, November 17-22,1991. y '
[
"Separation of Hexavalent and Trivalent Chromium from Supercritical Water Oxidation Effluents • Sam
Rolians, Lixiong Li, and Earnest F. Gloyna, presented at the I & EC Special Symposium for the
American Chemical Society, Atlanta, GA, September 21 -23,1992. ;
'Kinetics and Reaction Pathways of Pyridine Oxidation in Supercritical Water," Neil Grain, Saadedine
Tebbal, Lixiong Li, and Earnest F. Gloyna, Industrial and Engineering Chemistry Research
32(10), 2259-68, 1993. '
"Behavior of Chromium During Supercritical Water Oxidation," Sam Rolians, MS Thesis, Department of
Civil Engineering, The University of Texas at Austin, May 1993,
"Supercritical Water Oxidation and Hydrolysis Kinetics of 2,4-Dichlorophenol and Pyridirie," Neil Grain
MS Thesis, Department of Civil Engineering, The University of Texas at Austin, May 1993.
"Supercritical Water Oxidation of Four Selected Priority Pollutants," Anusuya Kanthasamy, MS Thesis
Department of Civil Engineering, The University of Texas at Austin, May 1993. | '
-" • -" "i*
Presentations at SRP Spring Conference, April 25,1991. c" i
Presentations at SRP Fall Conference, October 16,1991.
Presentations at SRP Spring Conference, April 14,1992.
Presentations at SRP Fall Conference, October 14,1992.
72
-------
B. Photographs of Experimental Setups
Photo B.I Batch SCWO Reactor System #1
73
-------
Photo B.2 Continuous-Flow SCWO Reactor System #1
74
------- |