Thermodynamic considerations the application of reverse mode gasification to the destruction of hazardous substances

Article abstract of DOI:10.1021/es9812352

Previous studies by us have demonstrated the effectiveness of reverse mode gasification using a granular char matrix for treatment of hazardous wastes Calculations pertaining to this gasification are presented, including a one-dmansional temperature profile and a thermodynamic analysis. Equilibrium compositions were calculated by free energy minimization using commercially available software The calculated results were compared with experimental data for gasification of mixtures containing water, selected hydrocarbons, and used motor oil. Batch and continuous feed reactors were used with optimized operating parameters 10 generate the data The dry gas product obtained from gasification of water and selected hydrocarbons contains carbon dioxide, carbon monoxide, methane, and hydrogen, in agreement with thermodynamic predictions, and the compositions agree well with predictions obtained assuming that chemical equilibrium is attained at a temperature of 650 C The dry gas product from gasification of motor oil contains small amounts of low molecular weight hydrocarbons, which are not thermodynamically stable, but the composition of the major products generally agrees with the thermodynamic predictions. Under optimized conditions, the aqueous condensate contains between 1 and 100 ppm organics. Heat balance terms for the process were also calculated, and these demonstrate the efficiency of gasification as a treatment method.

Full text of DOI:10.1021/es9812352

Environ. Sci. Technol. 1999, 33, 2973-2979
1
. A granular subbituminous coal char with a macroporous
Thermodynamic Considerations in
the Application of Reverse Mode
Gasification to the Destruction of
Hazardous Substances
surface is used to sorb the waste. If required, water is added
to facilitate gasification. The details for preparing the char
and its sorption properties have been presented elsewhere
(
2, 4-8). After sorbing the waste, the char remains a dry
granular solid, easy to handle, so that problem materials
such as oily sludges can be readily processed. The char/
waste mixture is charged to the feed hopper and subsequently
into the cylindrical reactor. The granular char serves as a gas
permeable support matrix for the process. Heat is applied
D A V I D W . L A R S E N * A N D
M I C H A E L D . W A S H I N G T O N
at the ignition point, and O
reactor. After a short time, an incandescent thermal zone
ITZ) forms spontaneously at the point of heating, at which
2
gas is fed in at the top of the
Department of Chemistry, University of MissourisSt. Louis,
St. Louis, Missouri 63121
(
time heating is no longer required.
S T A N L E Y E . M A N A H A N A N D
B R A D L E Y M E D C A L F
The ITZ is a narrow, high-temperature zone, typically less
than 1 cm high, but spread evenly across the entire cross-
section of the reactor. In batch mode gasification, the ITZ
passes upward, once through the entire length of the char
bed, against the gas flow. In continuous operation, as shown,
the upward ITZ movement is balanced by the downward
movement of the char bed, as controlled by the valve, so that
the ITZ remains stationary within the reactor. Within the
ITZ, oxidative chemistry occurs with release of a large amount
of heat, giving rise to localized temperatures up to an
estimated 2000 °C. Visually the zone appears “white hot”.
Reactions occur on the surface of the char and in the gas
phase. The heat of reaction is supplied primarily by the
oxidation of organics in the waste, and little of the char is
consumed. The region immediately above the ITZ is at
ambient temperature. Immediately below the ITZ, the
temperature is about 900 °C, and the char has a dull red
appearance. At the bottom edge of the ITZ and below, oxygen
is depleted so that the hot bed is both highly reactive and
chemically reductive. This facilitates reductive endothermic
chemistry and formation of the combustible gas mixture that
is characteristic of gasification. The endothermic processes
and heat loss cause continued cooling of the bed, and the
temperature decreases in the vertical direction to about 500
Department of Chemistry, University of MissourisColumbia,
Columbia, Missouri 65211
F R A N K E . S T A R Y
Department of Chemistry, Maryville University,
St. Louis, Missouri 63141
Previous studies by us have demonstrated the effectiveness
of reverse mode gasification using a granular char
matrix for treatment of hazardous wastes. Calculations
pertaining to this gasification are presented, including a one-
dimensional temperature profile and a thermodynamic
analysis. Equilibrium compositions were calculated by free
energy minimization using commercially available
software. The calculated results were compared with
experimental data for gasification of mixtures containing
water, selected hydrocarbons, and used motor oil. Batch and
continuous feed reactors were used with optimized
operating parameters to generate the data. The dry gas
product obtained from gasification of water and selected
hydrocarbons contains carbon dioxide, carbon monoxide,
methane, and hydrogen, in agreement with thermodynamic
predictions, and the compositions agree well with
predictions obtained assuming that chemical equilibrium
is attained at a temperature of 650 °C. The dry gas product
from gasification of motor oil contains small amounts of
low molecular weight hydrocarbons, which are not
thermodynamically stable, but the composition of the
major products generally agrees with the thermodynamic
predictions. Under optimized conditions, the aqueous
condensate contains between 1 and 100 ppm organics.
Heat balance terms for the process were also calculated,
and these demonstrate the efficiency of gasification as
a treatment method.
°
C at the bottom of the reactor.
The product gas leaving the bottom of the reactor passes
through a condenser, from which the dried gas emerges clean-
burning, with a moderate fuel value (e.g., 300 Btu/ (standard
3
ft ). The overall result is that the organic and aqueous portions
of the waste are removed from the char, with only a small
amount (e.g., 10%) of char consumption. The recovered char
is thus regenerated and can be recycled. It also serves as a
trap for undesirable byproducts.
Tem perature Profile. A one-dimensional theoretical
analysis of the complex temperature profile for underground
coal gasification derived from the heat equation has been
presented (14). The treatment is generally applicable to the
gasification studies described herein. Heat balance on the
element includes terms for heat transfer (by convection,
conduction, and heat loss), mass transfer (flow of gases),
chemical reaction kinetics, and reaction enthalpies. A single
Arrhenius equation for the rate and separate reaction
enthalpies for the exothermic and endothermic processes
were used for simplicity. The resulting differential equations
require numerical solution. The predicted temperature profile
using parameters applicable to underground coal gasification
(14) is shown in Figure 2. The temperature profile represents
a front that moves along the bed (or reactor) axis at a fixed
velocity, depending on the values of the operating parameters;
movement is from right to left in the figure. The ITZ is at the
temperature maximum, denoted by the dashed line. The
coal (or untreated feedstock) resides at the left, in the region
Introduction
We have shown (1-12) that “reverse mode” (13) gasification
which uses a specially prepared char support matrix is
convenient and effective in treating a variety of hazardous
substances. Reverse mode gasification as a continuous
process can be run using the configuration shown in Figure
*
Phone: (314) 516-5341; fax: (314) 516-5342; e-mail: larsen@jinx.
umsl.edu.
1
0.1021/es9812352 CCC: $18.00

1999 Am erican Chem ical Society
VOL. 33, NO. 17, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2 9 7 3
Published on Web 07/29/1999

FIGURE 1. Schematic of apparatus used for continuous feed reverse
mode gasification.
FIGURE 3. Calculated composition of product gas in reverse mode
gasification. The composition depends strongly on the thermal
equilibrium temperature as shown.
Thermodynamics provides the driving force, but kinetic
factors may control the chemistry, especially since the
residence time within the ITZ is relatively short. A thermo-
dynamic analysis of the process is given below. In addition,
studies to optimize process parameters are described.
Finally, as a first attempt to determine whether kinetic
limitations exist, the observed composition of the gas product
is compared with the predicted equilibrium composition.
Therm odynam ic Analysis. An important thermodynamic
consideration in gasification of organics and water on char
is the temperature-dependent equilibrium distribution of
products. This can be calculated with readily available
computer software (15). To illustrate, consider a simple model
system comprised of a char containing octane (representing
the organic content of a waste) and water which is gasified
using oxygen. The composition, which is typical for reverse
FIGURE 2. Temperature profile for reverse mode gasification. The
one-dimensional model represents a thermal front that passes
through a porous carbonaceous bed (see text for further details).
The parameters used to generate the plot are appropriate for
underground coal gasification.
mode gasification, is 1 mol of C
8
H
18(l), 38.5 mol of C(s), 6.4
mol of H O(l), and 5.8 mol of O
2
2
(g). This is a heavily loaded
char, organics and water each being 25% of the char weight.
The char is assumed to be graphite, denoted as C(s) (16), and
any of the large number of compounds in the database can
be included in the computation. After the starting composi-
tion is specified, the equilibrium composition of this system
can be calculated at any desired temperature and pressure.
Thermodynamic data from the database (17) are used, and
the total free energy of the system is minimized (18). The
results are shown in Figure 3, where the gas-phase product
distribution is plotted as a function of temperature, at 1 bar
pressure. It can be seen that the results are very simple; there
are only five thermodynamically stable product gases. The
marked “heating”, where the temperature increases rapidly
and exponentially as the ITZ moves. Movement is caused by
conduction of heat into the oxygen-rich region ahead (left)
of the ITZ, which causes the reaction rate to increase with
increasing temperature until the oxygen is completely
depleted at the point of maximum temperature. Gasified
char resides at the right, in the oxygen-depleted region
marked “cooling,” where the temperature falls off more
gradually with movement of the ITZ. In the original treatment
(
14) the rate of thermal decay in this region was attributed
composition is temperature-dependent; CH
are predominant at low temperatures, and the high-tem-
perature products are CO and H . The ratio of the products
depends on the starting composition of the system, but the
same gases and the same qualitative trends are always
predicted regardless of the hydrocarbon present initially.
Oxygen-containing organics are also converted to these five
gases. Of special interest, it can be seen that if water is
removed with a condenser, the remaining gases should have
a moderately high Btu content at temperatures above about
4 2 2
, CO , and H O
solely to heat loss. We have modified the equations to also
include the effects of endothermic chemistry, which also
contribute to cooling, as shown. Quantitatively, Figure 2
applies strictly to coal gasification, but the qualitative features
are the same as those in the present studies, so the figure is
of use in visualizing the gasification process. The quantitative
differences are attributable to the present use of oxygen, as
opposed to air in coal gasification, to the neglect of radiative
heat transfer (which is not as important in coal gasification
with air but which should be important in the present studies)
and to the differences between waste laden char and coal.
2
6
00 °C.
C(s), not shown in the figure, is also present in large excess
across the entire temperature range. At low temperatures,
near 200 °C, additional C is formed (about 10% excess). C(s)
gradually decreases as the temperature increases; about 5%
The one-dimensional gasification model, even though
oversimplified, is still complex.
2
9 7 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 17, 1999

is consumed in the intermediate temperature range, near
00 °C. In the high-temperature range, C(s) decreases further
6
so that about 25% is consumed at 800 °C, and above this
temperature, no further C is consumed.
If the results shown in Figure 3 are extended to higher
temperature, radicals are predicted; e.g., at 2000 °C, the
estimated temperature within the ITZ, one obtains 0.02 mol
-
3
-6
3
H, 10 mol CH , and 10 mol OH. The efficient destruction
of refractory organics, e.g., chlorinated compounds, has been
attributed especially to the presence of H (8).
The fate of heteroatoms can be similarly determined. Two
common ones of interest are S and N because of the potential
x x
formation of SO and NO . This problem does not occur with
reverse mode gasification as described herein. The stable
sulfur-containing species predicted during gasification are
FIGURE 4. Diagram illustrating thermodynamic heat balance terms
in reverse mode gasification.
S, COS, and H
2
S (19), and the stable nitrogen-containing
(11).
species are N , HCN, and NH
2
3
From Figure 3, the equilibrium product composition at
a given temperature and 1 bar can be determined. In general,
these data illustrate the thermodynamic driving force for the
process. However, they also allow quantitative predictions
to be made if one assumes that the system comes to
equilibrium at some temperature. The results of our previous
studies generally support this approach. First, the gas product
is combustible, and within certain ranges of operating
parameters, it is possible to get a clean-burning gas product,
i.e., one that burns with a blue flame. This indicates that
larger molecular weight hydrocarbons are present only at
low levels. The gas composition suggests that the temperature
for equilibrium is between 600 and 700 °C. The limits of
combustibility of gases in air (20) are as follows: hydrogen,
A second important thermodynamic consideration is heat
balance in gasification. This can be discussed in terms of the
process diagram shown in Figure 4. Gasification per se as it
occurs within the reactor is described by the vertical process
on the left, where Tex is the exit temperature of the gas product
and ∆exH is the reaction enthalpy, which is the heat lost to
the surroundings during the process. Under adiabatic
a
conditions, i.e., when ∆exH ) 0, Tex equals T , the adiabatic
temperature, which is the highest temperature attainable at
the exit port of the reactor. The sensible heat, -qsens, is that
which must be removed from gas and char products to reduce
g
the temperature from Tex to 25 °C. ∆ H(25 °C) refers to
gasification in the standard state. Finally, burning the gas
mixture releases the (standard state) heat of combustion,
4
-74%; methane, 5-15%; carbon monoxide, 12.5-74.2%.
∆
c
H(25 °C). Idealized incineration produces complete com-
bustion products (CO and H O) with corresponding standard
state ∆ H(25 °C). The unique features of gasification in
Inspection of Figure 3 shows that, for temperatures above
about 400 °C, the equilibrium gas product mixture should
sustain combustion. Second, water is present in the gas
product, and it is conveniently condensed and recycled in
the process. This indicates an equilibrium temperature below
2
2
i
contrast to incineration can be demonstrated numerically
with reference to Figure 4.
Before determining the adiabatic temperature, it is useful
to consider the gas-phase reactions per se. These strongly
exothermic processes lead to very high temperatures in the
gas phase, which would be characterized in the limit of no
vapor-phase heat loss, by an adiabatic flame temperature
8
00 °C. Third, char loss during gasification is typically 10%
or less, which also suggests an equilibrium temperature below
about 700 °C. Temperatures within the ITZ are markedly
higher than these values; in fact to ascribe a temperature to
the ITZ is an oversimplification since exothermic gas-phase
reactions are the primary source of heat, and the heat must
then flow to the interior of the char particles. Direct
measurements with high-temperature thermocouples in-
serted in the ITZ have yielded readings of 1400 °C, but actual
gas-phase temperatures are probably higher; we estimate
(
21), which is typically >2000 °C. This high temperature is
consistent with the observed ITZ and the presence of radicals
as discussed above. However, we will consider the adiabatic
temperature to be that in which heat is allowed to flow to
the char. Since char is present in large excess, the resulting
adiabatic temperature will be lower than the adiabatic flame
temperature.
2
000 °C. Endothermic processes in the reduction zone cause
the temperature to fall rapidly as the gas moves through the
reactor so that the gas normally exits the reactor at about 500
a
T for the model system described above was found with
the use of the thermodynamics software (22) to be 714 °C.
This value serves as a check on the calculation because it
must be above the observed exit temperature, which is about
°
C. The magnitude of these temperatures indicates that
establishment of equilibrium must occur in the reductive
zone. Last, SO and NO in the gas product are generally
x
x
5
00 °C. Next, we consider a nonadiabatic gasification. It is
unobservable, also in accord with the thermodynamic
predictions.
necessary to specify an equilibrium temperature for the
products, which will determine the composition. We will
show below that a reasonable value for the equilibrium
temperature is 650 °C. A typical experimental exit temper-
ature, Tex is 500 °C. The calculated enthalpy values for heat
balance in gasification are then
The quantitative application of these data to gasification
is not obvious because of the complex temperature profile
and also because the kinetics of the chemical reactions are
undoubtedly complex. However, the equilibrium predictions
can be considered with the following rationale. First, the
temperature decreases rapidly as the gases leave the ITZ and
pass through the reductive or cooling zone. Second, the rates
of reaction are highly temperature-dependent, and these will
decrease markedly along the axis of the reactor. Thus, we
consider that, at some position, there is a cutoff temperature,
below which the reaction rates are negligible so that
equilibrium is established at that temperature. As a test of
this approach, experimental gas compositions can be com-
pared with the thermodynamic predictions to determine
whether agreement can be found at some temperature.
∆
exH
-qsens
∆gH(25 °C)
-1466 kJ
∆cH(25 °C)
-5321 kJ
∆iH(25 °C)
-6787 kJ
-819 kJ
-647 kJ
∆
exH, the heat lost to the surroundings during gasification,
is about 12% of ∆ H. The sensible heat qsens, which is that
contained by the products by virtue of these being at elevated
temperature, is about 10% of ∆ H. Both these quantities can
be partially recovered with heat exchangers. An important
i
i
VOL. 33, NO. 17, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2 9 7 5

measure of the efficiency of the process is ∆
c
H/ ∆
i
H, the ratio
good medium for gasification. The simplest system for
gasification is char containing a small amount of water. Thus,
as a starting point, we consider char containing 10% water
based on char weight. Gasification of this system does not
require optimization since it is relatively insensitive to oxidant
flow. Because of its simplicity, this should be the ideal system
with which to compare the experimental and calculated gas
compositions. Reverse mode gasification with a Vycor reactor
using oxygen gave the following dry gas product with
composition, as determined by GC:
of the heat released in the combustion of gas products to
that of incineration. In the present case the ratio is 78%, and
this illustrates very clearly how resource recovery, in the form
of fuel gas, can be produced along with destruction of
hazardous materials.
The thermodynamic basis for reverse mode gasification
is thus established. However, the analysis is subject to kinetic
limitations. We note that when this method is used for the
treatment of hazardous mixtures, the aqueous condensate
and char product can be conveniently recycled. This is
fortunate because both char and condensate serve to trap
hazardous substances while allowing synthesis gases to
emerge. In practice a char filter may also be placed at the
outlet of the condenser, and this char can also be regenerated
in the process. The most important requirement is that the
product gas be free from hazardous byproducts. In the
remainder of this report, we present results of several studies
to determine the extent to which equilibrium compositions
can actually be obtained in reverse mode gasification. We
also present thermodynamic analyses of the experimental
results, which broaden our understanding of the process.
gas component
H2
CO
CH4
CO2
obsd fraction
0.246
0.360
0.026
0.368
The chromatograms gave no indication of gases other
than the four shown, which is in agreement with thermo-
dynamic predictions. Standard deviations in the analyses
were about 1% of the reported values.
Computer-generated equilibrium compositions are shown
below for three equilibrium temperatures 600, 650, and 700
°
C:
Experimental Section
Reactors. Batch mode gasifications were conducted in a 1
in. diameter by 1 ft length Vycor tube and in stainless steel
tubes (1 and 1.75 in. diameter) by 2.5 ft. Continuous feed
gasifications were conducted with a 2 in. diameter stainless
steel reactor (as described in Figure 1).
equilib temp
H2
CO
CH4
CO2
6
6
00
50
0.198
0.212
0.215
0.236
0.373
0.519
0.015
0.010
0.006
0.551
0.405
0.260
700
Char. One type of char used in the studies was prepared
by gasifying coal multiple times in reverse mode. A second
type of char was prepared by pyrolyzing coal.
Gas Sam pling. Effluent from reactors passed through a
cold trap set in an ice bath, which served to remove
condensables. The trap was fitted with a sidearm for the
collection of the off gases. The gas samples were then collected
in 2 L Tedlar gas sample bags and immediately analyzed to
prevent reactions from occurring that might significantly alter
the gas composition.
The fractions for 650 °C agree reasonably well with the
experimental data. Fractions for 600 and 700 °C are markedly
different, which illustrates the sensitivity of the results to the
assumed temperature. There are large changes in CO and
, which suggests that the CO/ CO ratio is a good measure
2 2
of equilibrium temperature.
CO
The observed amount of methane is over twice what is
predicted thermodynamically. This may indicate that meth-
ane forming reactions continue to occur as the gas passes
through the lower temperature region of the reactor. Methane
formation is favorable at lower temperature; for example,
the equilibrium fraction of methane is 0.026 at 525 °C, and
the maximum equilibrium methane concentration is 0.041
at about 350 °C.
Condensates. Condensates were extracted with 3 mL of
deuterated chloroform prior to GC/ MS or NMR analysis.
Gas Chrom atography. Samples were analyzed for H
, and CO , using a Shimadzu Model GC-14A gas chro-
matograph with a carbosphere 80/ 100 column (Alltech, 6 ft
2
, CO,
CH
4
2
1
×
/
8
in.), a thermal conductivity detector, and an air-actuated
The amounts of char product and condensed water
recovered are also predicted in the calculations; however,
neither quantity can be readily determined experimentally.
This is because a layer of clean char is placed in the bottom
of the reactor to avoid volatilization of organics during startup.
Qualitatively, we observe a slight loss of char, about 10%,
and roughly the same amount of water is recovered as we
added initially. These observations are consistent with the
thermodynamic predictions.
1
mL gas sample loop. The oven temperature program was
3
5 °C for 2 min and ramped at 20 °C/ min to 155 °C. The
detector temperature and current were set to 155 °C and 125
mV, respectively. Helium was used as a carrier gas with a
flow rate of 25.00 mL/ min and a makeup flow of 20 mL/ min.
Gases were identified by comparing retention times with
standards of the pure gases. Calibration curves were produced
using the pure gas standards.
GC/MS. Measurements were made using a Hewlett-
Packard GC 5890 Series II gas chromatograph with a Hewlett-
Packard MS quadrapole engine MS 59827A. The GC was
equipped with a 0.32 mm i.d. × 30 m DB-5 column operated
at 50 °C for 4 min and then ramped 8 °C/ min to 300 °C. The
mass spectrometer scan covered the range from 50 to 800
Da. The Wiley/ NBS Registry of MS Spectral Data Library was
used for compound identifications.
Gasification of Char with an Enhanced Reductive Zone.
The previous results indicate that the equilibrium gas
temperature is about 650 °C. Since equilibrium presumably
occurs in the reductive zone, it should be possible to change
the composition of the gas product by varying the temper-
ature of the reductive zone. Increasing the temperature will
lower the heat loss, promote endothermic chemistry, and
increase the high-temperature residence time of the gas. The
net effect should be to shift the equilibrium toward more CO
High-Resolution Nuclear Magnetic Resonance. Nuclear
magnetic resonance (NMR) spectra were obtained with either
a 300 or 400 MHz Varian NMR spectrometer containing a
and H
was investigated by conducting two gasifications of char with
0% water, in which the region surrounding the reductive
2
, thereby increasing the Btu value of the dry gas. This
1
19
13
four-nucleus NMR probe (5 mm) for observing H, F, C,
and P.
1
3
1
zone was maintained first at 500 °C and second at 700 °C.
These temperatures bracketed the estimated equilibrium
temperature of 650 °C. Temperature control was accom-
plished by inserting the reactor in a tube furnace. The
Results and Discussion
Gasification of Char Containing Water. Dry char is not a
2
9 7 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 17, 1999

experimental dry gas compositions, as determined by GC,
are
A hydrocarbon-rich gas burns with a yellow flame. GC/ MS
analysis showed only trace amounts of oil in the condensate,
1
.9 ppm 1,3-dimethylbenzene, 8.3 ppm phenol, 17.5 ppm
hexadecane, and 1.5 ppm phenanthrene. Thus, the ther-
modynamic predictions for the condensate are realized to
within about 30 ppm of nonsynthesis gas components.
reductive zone temp, °C
H2
CO
CH4
CO2
5
7
00
00
0.209
0.252
0.341
0.576
0.017
0.015
0.434
0.158
Gasification with 3.5 L/ min (nonoptimized) oxygen flow
gave poor results, as evidenced by the yellow burning gas.
GC/ MS analysis of the condensate showed large amounts of
starting hydrocarbons (650 ppm decane and 560 ppm
hexadecane) as well as small amounts of substituted phenols,
naphthalenes, and other hydrocarbons, which indicates that
the shorter hydrocarbons tend to escape the reactor intact.
The calculated mole fractions in dry gas corresponding
to the observed data are
equilib temp, °C
H2
CO
CH4
CO2
6
7
40
40
0.210
0.213
0.344
0.618
0.011
0.004
0.435
0.165
Gasification of a Com plex Refractory Waste. Used motor
oil is a complex mixture that typically contains a fraction of
sludge, and this was chosen for study as being representative
of a real world waste. Complete characterization is eco-
nomically unfeasible, and its refractory nature makes it
difficult to destroy cleanly. While gasification of motor oil is
feasible, the results of our studies show that it must be done
with care. The first problem that may be encountered is
formation of an aerosol in the effluent gas, which was
particularly evident when gasification was done in a 1 in.
diameter Vycor reactor. The char contained 25%:25% oil/
water based on weight of char, and the oxygen flow was 3.5
L/ min. Under these conditions, an aerosol is clearly visible
in the gas product stream. The aerosol readily passes through
the condenser and also through a granular char filter placed
after the condenser. Some of the gas product was collected
for analysis, and the remainder was flared as it emerged from
the condenser through a tube. Under the conditions used,
the flared gas exhibits a blue flame when the higher
hydrocarbon content is low, and as the hydrocarbon content
increases, the flame becomes increasingly yellow. An oil layer
in the aqueous condensate is also observed in the latter case.
With more sophisticated flare design, it is possible to produce
a blue flame for most gas product compositions; however,
in the present studies, the blue flame was used as a simple
diagnostic test to determine proper operating parameters
for each gasification.
As observed in the previous experiment, the chromato-
gram indicated that only the four gases shown were present.
There is excellent agreement between the 500 °C reductive
zone results and the calculated results assuming a 640 °C
equilibrium. The observed results are similar to those
obtained previously without external heating of the zone.
The 700 °C reductive zone results are quite different, and
they correlate fairly well with calculated fractions at equi-
librium temperatures of 740 °C. This demonstrates that the
reductive zone processes are enhanced in this case. As
previously observed, the dramatic change in the ratio of CO/
2
CO is predicted and the hydrogen predictions are reasonably
accurate. However, as in the previous case, the amount of
methane predicted is below that observed and the predicted
decrease in methane with increasing temperature is also not
3
observed. The calculated Btu/ (standard ft ) values for the
observed gas compositions are 189 (500 °C) and 323 (700 °C).
Calculated heat balance terms in the reductive zone
experiments were found to be
reductive
zone temp, °C
∆exH,
kJ
∆gH(25 °C),
fractional
heat loss
kJ
5
7
00
00
-12.86
-1.49
-26.26
-17.54
0.490
0.085
The values expressed are per mole of char gasified. The
fractional heat loss in the last column is the ratio ∆exH/ ∆
25 °C), which indicates that there is very little heat loss when
With the Vycor reactor, about 15% of the condensate was
an oil layer and the aqueous layer contained an estimated
g
H
(
1
450 ppm dissolved organics, a combination of substituted
the reductive zone surroundings are held at 700 °C. The
calculation showed that the adiabatic temperature for
gasification of char containing 10% water is about 755 °C,
which is very close to the equilibrium temperature of 740 °C.
With a 500 °C reductive zone, the equilibrium temperature
is estimated to be 640 °C, well below the adiabatic temper-
ature, so that a substantial heat loss is obtained.
Gasification of a Char Containing Water and Hydro-
carbons. The ultimate objective of these gasifications is to
destroy complex waste mixtures containing organics. Thus,
as a next step, it was decided to study a mixture of equal
weights of decane, hexadecane, and octacosane, straight
chain hydrocarbons of varying length. Gasification of this
mixture will enable several factors to be determined. First,
to what extent can gasification conditions be optimized so
that only the synthesis gas components leave the reactor, in
which case, the condensate will contain only water? Second,
with nonoptimized oxygen flow rate, do these specific
compounds tend to escape the reactor intact or do they tend
to be converted to other substances. To study these questions,
we will focus on analysis of the condensate.
benzenes, naphthalenes, straight chain hydrocarbons, and
phenols. The presence of an aerosol during gasification is
highly undesirable since it can facilitate the release of
hazardous byproducts. The aerosol formation indicates
inefficient gasification, both in the breakdown of larger
organics in the ITZ and in their conversion to synthesis gas
in the reductive zone. It was determined that by use of a
larger diameter stainless steel reactor, the aerosol can be
eliminated, and we attribute this to reduced heat loss.
A second requirement for efficient gasification of motor
oil is sufficient water content in the char bed. A char sample
with 25%:0% oil/ water based on char weight was gasified
with an oxygen flow of 4.8 L/ min. The ITZ was difficult to
sustain and had to be reinitiated several times. The flared
gas product exhibited a yellow flame, and there was a 10%
oil layer in the condensate. Gasification of a sample with
2
5%:12.5% oil/ water at an oxygen flow of 2.5 L/ min gave a
sustainable ITZ but similar flame and condensate results.
After having conducted numerous gasifications, we conclude
that, as a rule of thumb, equal amounts of oil and water can
be satisfactorily gasified. However, the oxidant flow rate is
also important. This is illustrated with representative results
for gasification of 12.5%:12.5% motor oil/ water on char
A char containing 12.5%:12.5% mixture/ water based on
char weight was studied, and optimum results were obtained
with an oxygen flow of 4.8 L/ min. The gas burned with a blue
flame, which indicates that there is little hydrocarbon content.
VOL. 33, NO. 17, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2 9 7 7

obtained using a 1.25 in. stainless steel reactor:
TABLE 1. Composition of Gas Obtained with Continuous Feed
Reactor
sample 1
sample 2
2.1
sample 3
3.5
composition for given oxidant flow, %
oxygen flow, L/m in
gas flam e color
condensate weight, g 8.0
oil layer fraction, %
aq organics, ppm
4.8
7.9 L/min
8.7 L/min
10.5 L/min
blue
blue/yellow blue/yellow
hydrogen
propane
propylene
isobutane
n-butane
isopentane
n-pentane
carbon dioxide
ethylene
ethane
oxygen
nitrogen
m ethane
15.42
0.41
1.13
0.00
0.00
0.16
0.00
49.81
2.98
0.12
0.46
0.81
1.52
27.23
21.93
0.29
0.77
0.00
0.00
0.09
0.00
35.06
1.30
0.08
0.19
0.45
2.60
37.14
17.89
0.22
0.46
0.00
0.00
0.08
0.00
35.24
2.20
0.08
0.42
0.87
1.90
40.85
7.1
2
8.7
3
trace
<0.1
300
156
The results obtained for 4.8 L/ min oxygen flow were near
optimum. The flared gas exhibited a blue flame, and no oil
was visible in the condensate. The organic content as
determined by GC/ MS was very low. The only substance
found was phenol. Proton NMR showed very weak peaks
consistent with phenol. Gasification at other oxygen flow
rates produced blue/ yellow gas flares, oil layers in the
condensate, and greater aqueous organics. The compounds
identified by GC/ MS were benzenes, naphthalenes, an-
thracenes, straight chain hydrocarbons, and phenols.
carbon m onoxide
components, the majority being ethylene, propylene, and
propane, is indicative of nonequilibrium conditions. The
presence of oxygen and nitrogen is most likely caused by
Gasification of samples containing 25%:25% oil/ water gave
similar results:
2
leaks. There is also a marked shift in the CO/ CO ratio with
increased oxygen flow rate, which indicates a shift to higher
temperature equilibrium.
sample 4
sample 5
4.1
sample 6
4.8
oxygen flow, L/m in
gas flam e
3.5
blue/sl. yellow blue/yellow blue/yellow
condensate weight, g 11.3
The agreement between thermodynamic predictions and
observed gas-phase composition obtained with optimized
operating parameters indicates the absence of kinetic
limitations in the formation of gas products. This is also
consistent with the observed low organic content in the
aqueous condensate.
10.7
10
300
14.9
2
414
oil layer fraction, %
aq organics, ppm
trace
124
The optimum oxygen flow rate is near 3.5 L/ min. Thus,
in the composition range of these samples, the oxygen flow
rate seems to depend inversely on oil loading. The com-
pounds found by GC/ MS in sample 4 were mostly substituted
phenols with smaller amounts of naphthalenes and benzenes.
NMR results were consistent with phenol. Many compounds
were found by GC/ MS for samples 5 and 6. NMR spectra
showed large peaks attributable to methyl and methylene
groups, with only very small peaks downfield.
Literature Cited
(
1) Cady, C.; Kapila, S.; Manahan, S. E.; Larsen, D. W.; Yanders, A.
F. Chemosphere 1990, 20, 1959.
(2) Larsen, D. W.; Manahan, S. E. U.S. Patent 4,978,477, 1990.
(3) McGowin, A. E.; Kinner, L. L.; Manahan, S. E.; Larsen, D. W.
Chemosphere 1991, 22, 1191.
(
(
4) Larsen, D. W.; Manahan, S. E. U.S. Patent 5,124,292, 1992.
5) McGowin, A. E.; Kinner, L. L.; Manahan, S. E.; Larsen, D. W.
Chemosphere 1992, 24, 1867.
Gasification with a Continuous Feed Reactor. Another
factor in the destruction of motor oil is the anticipated
increase in throughput with use of continuous processing
with the type reactor shown schematically in Figure 1. The
gasification of motor oil was studied with such a reactor, and
its use afforded the opportunity to study some additional
factors of interest. First, the apparatus required for continuous
mode operation is more complex. The rates at which char
is fed into and removed from the reactor as well as the rate
of oxidant flow must be carefully controlled so that the ITZ
remains stationary. This control is facilitated by monitoring
temperature at several points in the reactor. Considerable
time was spent trouble shooting the apparatus. However
with a smoothly operating continuous mode unit, much more
material could be treated with each run and it was possible
to extend the time for each run to as long as 90 min. This
length of time gave the opportunity for operating parameters
to be varied and optimized.
(6) Kinner, L. L.; McGowin, A. E.; Manahan, S. E.; Larsen, D. W.
Environ. Sci. Technol. 1993, 27, 482.
(
(
7) Kinner, L. L.; Manahan, S. E.; Larsen, D. W. J. Environ. Sci.
Health 1993, A28, 697.
8) Kinner, L. L.; McGowin, A. E.; Manahan, S. E.; Larsen, D. W.
Toxicol. Environ. Chem. 1993, 37, 1.
(9) McGowin, A. E.; Brendt, W.; Manahan, S. E. Larsen, D. W.
Chemosphere 1993, 27, 807.
10) McGowin, A. E.; Cady, C.; Manahan, S. E. Larsen D. W.
Chemosphere 1993, 27, 779.
11) Medcalf, B. D.; Larsen, D. W.; Manahan, S. E. Environ. Sci.
Technol. 1997, 31, 194.
12) Morlando, R. A.; Larsen, D. W.; Manahan, S. E. Environ. Sci.
Technol. 1997, 31, 409.
(
(
(
(13) Berry, V. J.; Parrish, D. R. Petrol. Trans., AIME 1960, 219, 124.
(
14) The commercially available program HSC Chemistry was
developed by Outokompu Research Oy, Pori, Finland.
15) A typical analysis of char is 94.2% C, 5.5% H, and 0.3% O.
16) “HSC Chemisry” contains thermodynamic data for almost 10 000
compounds, including temperature dependent heat capacity
data. It is also possible to correct for nonideality of gases; this
was not done in the present case since these corrections are
expected to be small given the gasification conditions, 1 bar
and temperatures greater than 500 °C.
(
(
It was found that a relatively low loading of oil on the char
was required for prolonged smooth operation, e.g., a mixture
containing 5%:10% oil/ water based on char. The rate of
oxidant flow was varied, and the composition of the dry gas
was monitored by GC at the different flow rates as shown in
Table 1. There is larger experimental error for these data
which may be attributable to the mode of operation or to the
sampling procedure per se. However, the synthesis gas
compositions generally agree with thermodynamic predic-
tions. The presence of several percent of nonsynthesis gas
(
17) The mathematical basis for the calculation is given in e.g.: Kyle,
B. G. Chemical and Process Thermodynamics; Prentice Hall:
Englewood Cliffs, NJ, 1992; pp 431-436.
(
(
18) Medcalf, B. D.; Manahan, S. E.; Larsen, D. W. Waste Manage.
1
998, 197, 18.
19) Handbook of Chemistry and Physics; Chemical Rubber Publishing
Co: Cleveland, OH, 1955; pp 1770-1772, 37.
(20) Noggle, J. W. Physical Chemistry; Harper Collins: 1996; pp 281-
286.
2
9 7 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 17, 1999

(
21) The adiabatic process temperature is found by an iterative
procedure using HSC Chemistry. The initial reactant composi-
tion (char, water, octane, and oxygen), initial temperature, and
final (output) temperature are selected. With these data, the
product composition (char and gases) is calculated using HSC.
Next, initial and final temperatures and compositions are input
back into HSC, and the heat balance is evaluated. When the
final temperature equals the adiabatic temperature, ∆H ) 0.
Thus the adiabatic temperature can be found by iteration.
Received for review November 30, 1998. Revised manuscript
received May 26, 1999. Accepted June 16, 1999.
ES9812352
VOL. 33, NO. 17, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2 9 7 9