United States
Environmental Protection
Agency
Water Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-87/022 May 1987
&EPA Project Summary
Aqueous-Phase Oxidation of
Sludge Using the Vertical
Reaction Vessel System
This study provides plant-scale oper-
ating data on the wet oxidation of mu-
nicipal wastewater sludge by the Verti-
cal Reactor Vessel* (VRV) system. An
important consideration in the evalua-
tion was the effect of the return flow
from the wet oxidation process on the
operation of the wastewater treatment
plant. The investigation was carried out
at the Longmont, CO, Wastewater
Treatment Plant (WWTP).
The VRV system consists of a series
of long concentric tubes placed in the
earth using conventional oil field tech-
nology. The reason for using vertical
construction is to produce a large hy-
drostatic head at the bottom of the sys-
tem. The pressure is needed to prevent
boiling at the temperatures required for
wet oxidation. By utilizing hydrostatic
pressure, pumping is required only to
overcome frictional losses. The need to
add energy for pressurization is elimi-
nated. Sludge is introduced along with
air or oxygen into the multiphase fluid
downcomer which is the second con-
centric space from the outside. It is
pumped down this space where it is
heated by hot oxidized sludge rising in
the outermost concentric space within
the vessel. In the bottom of the vessel,
temperatures of 250°C or higher are at-
tained and the oxidation of organic ma-
terials takes place with resulting heat
production. At the center of the reac-
tion vessel is a tubular heat-exchange
system which can either extract excess
heat or provide heat for startup of the
process.
*Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
At temperatures above 260°C it was
possible to achieve total chemical oxy-
gen demand (COD) reduction of about
80% and total volatile solids reductions
of over 90%. For the 25-cm reaction ves-
sel installed at Longmont, the capacity
of the system using air was limited to 5
metric tons per day. Using oxygen it
was possible to increase capacity to
about 30 metric tons per day. Returning
the supernatant liquid from the process
to the wastewater treatment system
did not have a significant effect on the
effectiveness of that system.
This Project Summary was devel-
oped by EPA's Water Engineering Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Treatment and disposal of municipal
wastewater treatment plant sludge is
expensive, with the cost constituting as
much as one-half of the total cost for
wastewater treatment. In addition, dis-
posal of the sludge after treatment is
becoming an increasing problem in
many locations. There is growing resis-
tance to disposal in landfills because of
possible groundwater contamination,
and strong resistance to increased
ocean disposal from communities near
the coasts. Incineration, which accom-
plishes disposal except for a relatively
small volume of residual ash, is being
questioned because of possible air pol-
lution problems.
-------
Wet oxidation of sludge, which is sim-
ilar in some ways to incineration, pro-
duces a greatly reduced volume of a
material similar to ash. It has a signifi-
cantly reduced air pollution potential,
but does produce a liquid waste stream
containing low molecular weight or-
ganic materials, especially acids. This
liquid waste stream can be sent to the
mafn treatment system and does not
cause a disposal problem. Although the
technical feasibility of wet oxidation for
sludge treatment has been recognized,
the process is expensive and has not
been widely used. In the 1970's the con-
cept of carrying out wet oxidation in a
vertical reactor placed in the ground
was investigated. Using conventional
oil field technology, a very long reaction
vessel, over 1,000 m, could be con-
structed which would provide high
pressure at the bottom from the hydro-
static head. Sufficient pressure could be
developed to prevent vaporization of
water at temperatures above the 250°C
required to carry out sludge oxidation.
This vertical configuration eliminated
the need to provide energy for pressur-
izing the wet sludge. A system of con-
centric tubes was developed which pro-
vided an annulus for pumping the
sludge down the reactor and another
annulus where the sludge returned to
ground level. In addition, tubes were
provided for circulation of a heat-
transfer fluid down the reactor to add
heat during startup and to remove ex-
cess heat during operation. The eco-
nomics of this vertical configuration ap-
peared reasonable because of the
elimination of an expensive above-
ground pressure vessel, the need for
only a small land area, and reduction of
pumping energy to only that needed to
overcome wall friction.
A pilot test of the vertical configura-
tion was carried out in the late 1970's at
the Lowry Bombing Range near Denver.
The depth of the system was only about
460 m, and did not produce high
enough static pressure to prevent boil-
ing without pressurizing the system at
the surface. The results of this study
were sufficiently encouraging to test an
improved design at a municipal waste-
water treatment plant. The size of the
system needed to be sufficient to treat
all the sludge from the treatment plant.
The full-scale evaluation of a VRV sys-
tem was carried out at the Longmont
WWTP. The overall objective of this
study was to provide plant-scale operat-
ing data on sludge oxidation. Two
specific goals were:
To determine operating parameters
for the process which would result
in effective reduction of COD and
total suspended solids (TSS).
To determine the effects of the re-
turn of wet oxidation supernatant
on the biological treatment in the
Longmont system.
Procedure
A VRV system was sized to treat the
approximately 3,600 kg/day of sludge
produced by the Longmont plant and
was installed on the treatment plant
grounds. A diagram of the total sludge
treatment system is shown in Figure 1
and indicates the path of the sludge
through the reactor. The section of the
reactor containing heat transfer oil con-
sists of an outer and inner tube. Insula-
tion between the tubes reduces heat
transfer between them. The diameter of
the reactor is 25 cm and the length is
1,600 m. It was intended to operate the
system using air for oxidation, but this
procedure limited capacity and reduced
the effective hydrostatic head because
of the presence of relatively insoluble
nitrogen bubbles. A source of oxygen
was added to allow operation over a
range of oxygen contents up to 100%. A
Lamella Solids Separator, which is a va-
riety of tube settler, was used to sepa-
rate the oxidized, ash-like residue from
the wet-oxidation supernatant. Provi-
sion was made for returning the super-
natant to the wastewater treatment
plant.
The system was operated during
most of 1984 and 1985 under a variety
of operating conditions. COD removals
obtained are shown in Figure 2. Al-
though COD reduction is plotted as a
function of temperature, the effect of
other variables is also included and ap-
pears as scatter in the data. Ranges for
these variables are shown in the table
included on the figure. Figure 3 shows
volatile suspended solids reduction as a
function of temperature. Just as for the
COD plot, this plot includes data cover-
ing a range of other variables as indi-
cated by the table on the figure. These
results indicate that at temperatures
above 260°C very good reduction of
COD is obtained, and most of the
volatile suspended matter is both solu-
bilized and largely oxidized. The proc-
ess is, therefore, quite effective for pro-
ducing a low-volatile-content, ash-like
material. Return of the supernatant
from the oxidized product after solids
separation did not have a significant
negative effect in terms of effluent bio-
chemical oxygen demand (BOD) or sus-
pended solids content. There was a
slight increase in ammonia content in
the treatment plant effluent. The degree
of oxidation was not greatly affected by
the amount of excess oxygen available.
Good operation should be possible with
only slightly more than the stoichiomet-
ric amount required for COD reduction.
Conclusions
1. With the use of air, the system ca-
pacity was limited to 5 metric tons
per day for the Longmont 25-cm re-
action vessel. The use of oxygen in-
creased system capacity to a maxi-
mum of 30 metric tons per day.
2. For bottomhole temperatures of
260°C and above, COD reductions
of 75% to 80% were achieved and
reproducible. Temperature had the
largest effect on COD reduction. By
recycling the effluent, the reduction
in COD was increased about 5%.
3. BOD in the VRV effluent stream was
readily biodegradable, showing a
140% increase in the efficiency of
the trickling filter that served as a
roughing treatment upstream of a
rotating biological contactor. As a
result, the Longmont WWTP was
able to process all the recycled BOD
and still meet its NPDES discharge
limit, even during periods when ad-
ditional sludge was hauled in.
4. During autogenous operation, that
is, operation in which heat pro-
duced by oxidation was sufficient to
maintain the reactor temperature,
total volatile suspended solids re-
ductions averaged 96.5%, and total
volatile solids reductions averaged
93.9%. Total suspended solids re-
ductions averaged 78.7%, and COD
reductions averaged 76.3%.
5. An industrial hygiene survey indi-
cated that trace components in the
off-gas were usually about 2 orders-
of-magnitude below the permissi-
ble limit for an 8-hr worker expo-
sure.
6. A metallurgical inspection of the re-
action vessel by Material Science
Corp. showed insignificant corro-
sion, consistent with a 20-year life.
Materials of construction for the
VRV and the surface equipment
proved satisfactory.
7. Based on the Longmont demon-
stration, it was possible to establish
operating and maintenance costs
for a 25-cm reaction vessel operat-
ing at maximum capacity. Operat-
-------
From Atmos.
Nitrogen-*- . i
Feed f^ K.
(^ JWea
' * t
r
Sludge Thickent
?--£"--"--'
*.' -' ^i '""
V_y ^ 1
r*- |
1 J
Longmont \
* n
Sludge Feed -» cxj-LJ-
TrucA S/urffl
Connection'*'"] Grind
Process ~ t-j-
M/afer
Process ? o
Wa^er >J 'j
Effluent ^ \
Recycle
,^c
__,-
Sca/e
Inhibitor
CO » cxi
f Xlffenf Tan/r ^y compressor
xt~£ \m \ I
tea? Xlffenr f^ X ft '
^"'""^ nr^i ff8'6'' ' M * **
Heaf
Exchanger
reeu
Tank heed (S-3) . ; .
.Raw Sludge (S-1) \
t^* t-o-jT^"3 ["^"^^i*^
e S/odffe *J^" :
er Pump Feed Pump ;
Diluent (S-2)
\ M
r Diluent
? Pump . '
r - |
^ V^
flea
Ve
Gas to _ _
/i.^^o i Off -Gases
Effluent (S-4) Atmos^.
\ Gas Separator
I-, 1 1 -txj-4 >
' Stilling
Well
To
;. £fluent^^
I Recycle
_y i
cf/on
sse/
X
^Supernatant (S-5)
/* \
--\ Solids
\Separator) Underfh
^^^ £o/ids (S
Ash I
{ 1 S/arry
| 1 Pum» \ { {
\ \ \\ l-l
1 J±1.*J -"S/7 P/'fS ]
Sump Pump '
-*-t»*- Effluent
Hauling
Ash
Truck
Figure 1. Process flow diagram.
10,
11
ing and maintenance costs for a
Longmont-sized unit are below
$100 per metric ton with an energy
recovery credit of $17 per metric
ton.
The ash is nonhazardous.
Leachates from the EP Toxicity Test
were typically 2 orders-of-
magnitude below limits set by EPA.
Ash in the VRV effluent was easily
dewatered. Solid contents of 40% to
75% were obtained after dewater-
ing by a centrifuge without any opti-
mization of centrifuge operation.
The system operated from non-
autogenous to extraction of heat.
Operation in the autogenous mode
was extremely smooth. Wash-out
heat was limited to a temperature
rise between influent and effluent
of 11-17°C (20-30°F) during autoge-
nous operation.
High mechanical availability
(96.7%) of the system and its com-
ponents did not limit the test pro-
gram in any substantial way.
12. COD reduction for the full-scale sys-
tem at Longmont was successfully
predicted from laboratory batch re-
actor (LBR) tests.
13. VRV operation was consistent and
reproducible at fixed operating
parameters.
Recommendations
The following are recommendations
for improving system design, based on
the results of the demonstration period:
1. Biological treatment of the reaction
vessel effluent should be further in-
vestigated to optimize COD reduc-
tion. Benefits of anaerobic polish-
ing should be tested.
2. The VRV effluent return appeared to
influence the biological speciation
and caused more rapid reduction of
BOD. Identification of biological
species before, during, and after ox-
idation system operating periods
should be included in future tests.
3. Returning the VRV effluent stream
' to the headworks of the Longmont
WWTP would provide flow equal-
ization and reduce concentration
prior to undergoing biological
degradation. An option should be
provided to allow VRV effluent re-
turn either to the headworks or to
the trickling filters.
4. The unit installed at Longmont was
oversized. A new reaction vessel
more closely matching Longmont's
sludge generating capacity should
be installed and tested.
5. Higher influent heat transfer fluid
temperatures were required for
startup due to welding failures
which occured in four sections of
the insulated tubulars. Coupling de-
sign also contributed to greater
heat loss than expected. However,
the system was still able to reach
-------
;oor
35
90
SB
Of
80
75
70
65
6O
%Q
|55
'§50
835
°30
25
20
15
10
Q
Q
D "^ H D
«$? 'ftnt
f'fi ,
IP
LJ - " u k
D D cP D F) cP1
n tn
D
D
a
-, 3ij3°anjn|5nr
|ij"JHffjjSJTtjjl
pSP^IJibjppl
a l"^ ffie a
n o
a rj]
Pj^Jfc
] rBJD
no
Hh
n
° 0 D Operating Conditions
; Q
COD Influent, mg/L
COD Effluent, mg/L
0 COD Reduction %
Bottomhole Temp. C
Sludge Load, Ib/hr
Liquid Flow, gpm
Air Flow. Ib/min
Oxygen Flow, Ib/min
22O 225 230 235 24O 245 250 255 260 265
Bottomhole Temperature (C)
Figure 2. CODT reduction vs. bottomhole temperature.
100-f
95
90
oc
GO
80'
75-
70
65
60
St 55
,| 50
||45
» 40
co 35
to
£30
J.O
20
15
10
o
n
n r] a
n jfi^pfp3
cfl
en
a a
a
Average
21.400
5,030
76.3
267.5
1200.0
108.8
3.4
17.2
270 275
Sxi QjiLE
-fPl-raTb
a
fidnoG
(Low-High)
5,600-48,500
1,400-19,400
32.3-92. 1
228-282
0-1,930
65-145
0-24.0
9.5-25.4
280 285 290
S&j-a-n
TP""
required bottomhole temperatures.
Improved insulated tubular design
should eliminate failures in future
installations. The performance of
the insulated tubular has minimal
effect during autogenous opera-
tion.
6. Pressure measurement tubing
caused organic fouling and suf-
fered mechanical damage during
installation and operation. Organic
fouling limited the length of operat-
ing runs. This fouling is not antici-
pated to occur in a commercial re-
action vessel, since downhole
pressure measurements will not be
required for VRV operation.
7. Mechanical dewatering of the ash
slurry should be installed. Polymer
addition should be optimized to re-
duce dewatering costs.
8. The use of ash as a filler material for
brick manufacture was demon-
strated and could be implemented
at future installations where appro-
priate.
9. The VRV should be insulated or
constructed so as to reduce heat
losses to the formation. High-
boiling-temperature oil should not
be present in the annular space be-
tween the reaction vessel wall and
the primary casing string. These im-
provements will reduce initial heat-
up and time to restart after a shut-
down, and will increase heat
recovery.
The full report was submitted in fulfill-
ment of CS-809337-01 by The City of
Longmont, CO, under the partial spon-
sorship of the U.S. Environmental Pro-
tection Agency.
Operating Conditions
COD Influent, mg/L
COD Effluent, mg/L
COD Reduction %
Bottomhole Temp. C
Sludge Load, Ib/hr
Liquid Flow, gpm
Air Flow, Ib/min
Oxygen Flow, Ib/min
Heat Transfer Fluid, Ib/min
Average
12.700
2,700
69.0
267.0
700.0
111.0
11.7
8.7
310.0
235 24O 245 25O 255 260 265 270
flange
(Low-High)
2,200-32,400
900-10,900
50-83.9
247-280
150-1,500
93-136
0-20.2
0-22.0
0-470
275 280
Bottomhole Temperature (C)
Flgura 3. TVSS reduction vs. bottomhole temperature.
4
-------
-------
This Project Summary was prepared by staff of the City of Longmont, CO 80501.
Edward J. Opatken was the EPA Project Officer (see below).
The complete report, entitled "Aqueous-Phase Oxidation of Sludge Using the
Vertical Reaction Vessel System," (Order No. PB 87-170 320/AS; Cost:
$18.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA 22161
Telephone: 703-487-4650
For further information. Carl Brunner can be contacted at:
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
EPA/600/S2-87/022
------- |