-------
0.6 ft
0.5
1.0 1.5
Percent PVC Added
2.0
2.5
FIGURE 6. CORROSION RATE OF A106 STEEL AT
VARIOUS TEMPERATURES VERSUS
PERCENT PVC ADDED TO REFUSE
12
-------
0.7
1.0 1.5
Percent PVC Added
FIGURE 7. CORROSION RATE OF Til STEEL AT
VARIOUS TEMPERATURES VERSUS
PERCENT PVC ADDED TO REFUSE
13
-------
the corrosion rate somewhat more than did the 2 percent-addition.
However, there were at least two differences in the experimental
conditions between Run 17 (2%) and Run 18 (170) :
(1) The incinerator operated 300 F hotter for
Run 18.
(2) Probe 17 was exposed to the hot gas of the
incinerator for <~-n 20 minutes before an KC1
detector registered an increase due to the
PVC loading.
An evaluation of these variables was made during the run with
Probes 19 and 20, the results of which are shown in Figures 8 and 9,
The major effect of incinerator gas temperature, which was 1400 F
for Probe 19 and 1550 F for Probe 20, on corrosion rate was found
to occur with specimens whose temperature exceeded 800 F. The
150 F "hotter gas caused a marked increase in corrosion rate and
the difference increased with increasing specimen temperature for
both the A106 and the Til steel. This increase in corrosion rate,
because of the hotter incinerator gas, was observed also for the
stainless steels as described in the next section. It is evident
from Figure 9 that preoxidizing had no significant effect on the
corrosion rate of the Til steel specimens on Probes 19 or 20.
Stainless Steels
All of the high-alloy stainless steels evaluated exhibited much
lower corrosion rates than the low-alloy steels even though the
stainless steels were, in general, exposed at higher specimen
temperatureso There was, as shown in Figures 10 and 11, essentially
no effect of specimen temperature or PVC content of the refuse on
the corrosion rate for the stainless steels. However, their
corrosion rates increased by a factor of 2X-6X when the incinerator
gas temperature increased. The effect of gas temperature on the
corrosion rates of the stainless steels is shown in the following
tabulation:
Alloy Average Corrosion Rate, mils/hr
No. 1400 F Gas > 1500 F Gas
310 0.032 0.170
446 0.050 0.136
825 0.056 0.259
316 0.056 0.203
321 0.109 0.201
14
-------
0.6
JC.
CO
0.5
0.4
0.3
0.2
O.I
Legend
X Probe 19
O Probe 20
200
400 600 800
Metal Temperature, F
1000
1200
FIGURE 8. EFFECT OF GAS TEMPERATURE ON
CORROSION RATE OF A106 STEEL
15
-------
.c
\
tn
Legsnd
X Probe 19
O Probe 20
D Preoxidized 4- hour at 1150 F in air
0.
200 400 600 800 1000 1200
Metal Temperature, F
FIGURE 9. EFFECT OF GAS TEMPERATURE ON
CORROSION RATE OF Til STEEL
16
-------
0.30
321 SS
Gas temperature >I550 F
0.20
D
0.10
Gas temperature 1400 F
A
X
a)
a
o
•r-l
(0
o
M
M
O
O
0
3 less
0.30
0.20
0,10
Gas temperature > 1550 F
Gas temperature 1400 F
Legend
X - 0% PVC
A,V - 0.5% PVC
D - 1.0% PVC
O - 2.0% PVC
O
-X.
D
310 SS
0.20
0.10
Gas temperature > 1550 F
T7
Gas temperature 1400 F
I
.n.
D
O
"X-
0
900
1000 MOO
Metal Temperature, F
FIGURE 10. CORROSION RATE OF ALLOY INDICATED
VERSUS TEMPERATURE
1200
17
-------
_c
01
to
OS
c
o
1-.
o
o
0.30
825
0.20
0.10
446
0.30
0.20
0.10
600
Gas temperature > 1550 F
Preoxidized specimen
D
Gas temperature 1400 F
O
J : 1 _
Legend
X - o% PVC
A V - 0.5% PVC
'a - 1.0% PVC
O - 2.0% PVC
Gas temperature > 1550 F
Preoxidized
specimen
n
o
Gas temperature 1400 F
*
D
x
800 1000
Metal Temperature, F
1200
FIGURE 11. CORROSION RATE OF ALLOYS INDICATED
VERSUS TEMPERATURE
18
-------
The major mode of attack is intergranular with some pitting
corrosion and grain fallout which produces a grainy (sugar-like)
surface. Preoxidation of Alloys 446 and 825 for 1/2 hour in air
at 1150 F did not significantly alter the mode or the extent of
attack, see Figure 11. Furthermore, no systematic differences in
corrosion rates cf the high alloy steels were observed that could
be related to the amount of PVC added to the refuse. If the heat
transfer tubes are located in an incinerator gas whose temperature
does not exceed 1400 F, the corrosion rate of the stainless steels
would be expected to be 1/6 to 1/10 that of the low-alloy steels
depending upon metal temperatures maintained. The high-alloy
steels do not suffer as severe a loss in corrosion resistance
with increasing metal temperature as is observed with the low
alloy steels. However, all steels investigated are very sensitive
to incinerator gas temperature when the specimen temperature
exceeds 800 F.
Corrosion Product and Deposit Analysis
As the corrosion probes were inserted into the fireside chamber of
the incinerator above the effluent end of the grate, the airflow
carried sufficient particulate matter to produce a deposit of 1/4
to 1/2-inch thick over approximately 1/3 of the 1-1/4-inch-diameter
tube in the 10-hour exposure periods. The remaining 2/3 of the
surface contained only a thin deposit and an oxide layer, the oxide
layer extending completely around the tube beneath the deposit.
The thickness of the oxide layer varied for the various alloys and
depended to a lesser extent upon specimen temperature. These
layers of deposit and oxide were removed mechanically and collected
by temperature zones for optical emission spectroscopy analysis of
the metallic element present. The results of optical emission
spectroscopy analyses are presented in Table 2. Also included in
Table 2 are the results of wet chemical analyses for chlorine and
sulfur concentrations and of phase analyses by X-ray diffraction.
The chloride concentration in the deposits decreases rapidly with
increasing specimen temperature as shown in Figure 12 f.or the five
probes, presumably due to reaction of the chloride with the furnace
gases and vaporization. It is interesting to note that the chloride
content of the deposit on specimens held at temperatures above 800 F
appears to increase linearly with the amount of PVC added to the
refuse as shown in Figure 13. However, at the lower specimen
temperatures, the chloride in the deposit increases more rapidly
between the 0 and C.5 percent PVC addition than between 0.5 and
2.0 percent addition. Furthermore, at the lower specimen temperature,
the amount of chloride in the deposit is lower on probes exposed to
the higher temperature (> 1550 F) incinerator gases than on probes
exposed to 1400 F gas. Thus, the data points, in parenthesis in
Figure 13, fall well below the joined data points.
19
-------
TABLE 2. RESULTS OF CHEMICAL ASSAY OF DEPOSITS
Elements
Detected
Na
K
Ca
Mg
Pb
Zn
Fe
Al
Si
Cl"
S=
02
Probe 16
400 F -
800 F
7.5
11.5
6.0
2.0
2.5
6.0
5.0
14.5
4.5
4.0
8.0
Ba lance
800 F -
1200 F
7.5
19.0
7.5
3.0
1.0
6.5
6.0
12.5
5.5
1.5
9.0
400
800
5.0
8.5
7.5
2.5
2.0
6.0
6,5
7.5
6.0
11.5
6.5
Probe 17
F - 800 F -
F 1200 F
Conceht'ra'
4.0
8.0
10.0
3.5
0.5
4.5
6.0
8.5
6.5
4.5
5.5
Probe
400 F -
800 F
tion, weight
6.5
10.8
10.8
3.3
3.0
6.5
8.5
6.5
7.5
5.8
6.9
-___
18
800 F -
1200 F
percent
4.5
7.5
12o5
3.0
0.3
4.5
6.3
7.3
7.5
1.3
8.6
Probe
400 F -
800 F
6.9
3.7
6.4
1.0
3.2
6.9
12.3
3.0
3.4
7.6
7.8
____
19
800 F -
1200 F
7.5
3.6
12.5
2:0
0.3
7.5
7.1
5.6
7.5
3.6
8.3
400
800
6.9
4.4
6.7
0.6
3.1
6.9
12.6
3.3
3.9
4.8
4.9
Probe 20
F - 800 F -
F 1200 F
7.5
3.7
12.5
2.0
0.3
7.5
6.6
6.0
7.5
1.9
7.6
CaSO,
Phase Studies on Deposits
CaSO, CaSO,
4 4
CaSO
CaSO
CaSO.
CaSO,
SiO
NaCl NaCl
KC1
sit>
NaCl
SiO,
-------
Legend
X - 0% PVC
A V - 0.5%
1.0%
200
400 600 800
Metal Temperature, F
1000
1200
FIGURE 12. CHLORIDE CONCENTRATION IN DEPOSIT ON
PROBES VERSUS METAL TEMPERATURE
21
-------
16
14
Legend
() Probes exposed to high temperature gas (>I550 F)
[] Probes exposed to high temperature gas H400 F)
Underl.ned temperatures indicate metal temperature
[X] 400 F
12
0.5 1.0 1.5
Percent PVC Added to Refuse
FIGURE 13. CHLORIDE CONCENTRATION IN DEPOSIT
VERSUS AMOUNT OF PVC ADDED TO REFUSE
22
-------
The sulfur content of the deposit is seen, in Table 2, to be more
uniform and perhaps slightly higher at the higher metal temperatures.
Several specimens representing various temperature zones were
reserved for X-ray diffraction phase analysis of the deposits and of
the scale layers between the oxide and the substrate metal. Phase
studies of the deposits are presented in Table 2 while the phase
studies on the scale beneath the oxide layer are presented in
Table 3. The oxide layer contains a mixture of Fe304 and F6203
which forms a semiperraeable barrier layer between the deposit and
the scale. The results presented in Table 3 on phases formed
between the substrate metal and the oxide layer show the mechanism
of attack to be similar to those presented in our summary report
under the previous (EP000325) grant. The major difference in the
low-temperature zone (400-800 F) is that the ferrous chloride forms
a continuous layer when PVC is added to the refuse while normal
refuse produces a discontinuous layer on the metal substrate. This
increase in the amount of ferrous chloride formed when PVC is added
to the refuse also contributes to the corrosion in the high-temperature
(800-1200 F) zone. Here the chloride melts, agglomerates, and
probably vaporizes exposing clean metal for the sulfide reaction.
During these changes in the chloride phase, the semiprotective
oxide layer on low alloy steel is undoubtedly partially destroyed.
However, the alloy oxide that forms on the stainless steels is
more resistant to this chloride attack.
Incinerator Gas Composition
Previous sections of this report have discussed the effect of refuse
composition on the corrosion of metal and the role of deposits and
scale upon the corrosion mechanisms. The composition of the deposits
is partially controlled by the noncombustible portion of the refuse,
by the gaseous species in the incinerator, and by the temperature of
the substrate and of the surrounding incinerator temperature. At
probe Location 1 of Figure 3 and to some extent at Location 2, the
off gases of the refuse are diluted with the overfire air, the
amount of which varies to maintain a temperature of 600 F at the
ID fan. As a result of variations in the amount of overfire air
the composition of the gas will fluctuate.
The composition of the refuse varies resulting in some changes in
off gas composition when refuse samples are combusted in the laboratory.
As shown in Table A-2 of the Appendix, the percent chlorine in the pff
gas from grab samples of refuse differs by a factor of over 2X
(0.347o to 0.74%). Table 4 shows fluctuations in gas composition for
Probe Runs 16, 17, and 18 where gas samples were collected in two
batches during each exposure period. Because of these wide
fluctuations, the gas was sampled throughout Run 19/20. As would
be expected, however, the amount of HC1 in the incinerator gas
(0.03 to 0.09%) is found to be 1/7 to 1/12 that evolved from the
refuse due to dilution by over- and under-fire air.
23
-------
TABLE 3. PHASE STUDIES OF SCALE NEXT TO METAL
Probe
Number
16
17
18
19
20
Phases By Layers Between the Metal and the Oxide
FeCl2
FeCl2
As 17
FeCl2
As 19
4CO F - 800 F
(*)
• 4H 0 - FeOOH - FeS
• 4H 0 - KC1
- no evidence of sulfide layer
(*•)
• 2H 0 - Fed • 4H20 - FeOOH - FeS
8CO *'
(FeCl2 • 4H2
(FeCl2 • 4H2
As 17(t)
(FeCl2 • 2H2
(FeCl2 • 2H2
Layer
- 1200 F
(*}
0 + FeOOH + FeS)v '
0 + FeOOH + NaCl)^
/... \
0 + FeOOH + FeS)V ;
(*)
0+FeSr '
(*) Phases in this temperature zone do not form successive layers but are
present in discrete areas.
(t) Analyzed by optical method.
This FeS is found at the interface between the ferrous chloride and
the oxide that separates the deposit from the scale.
-------
TABLE 4. INCINERATOR GAS ANALYSES TAKEN DURING CORROSION RATE STUDIES
ro
Ul
Probe Run Number
16(*>
Component
HC1,
HC1,
SO
SO
SO
SO
NO
NO
CO
°2
H2
2'
2'
3'
V
x'
x'
2'
ppm
ppm
Ppm
ppm
ppm
ppm
ppm
ppm
measured
corr. 12% CO
measured
corr. 12% C02
measured
corr. 12% CO
measured
corr. 12% CO
percent
, percent
o,
percent
1
289
1650
185
1850
2
20
35
200
2.1
18.7
12.3
2
458
4580
29
580
25
500
7
140
1.2
17.8
4.8
17 <*>
1
545
1090
56
168
4
12
73
219
6.0
15.2
6.9
2
1180
4720
82
328
14
56
38
152
4.0
17.2
8.0
1
548
1644
88
264
21
63
70
210
4.0
17.5
10.0
18(*> 19-20(t)
2 12
1260 583
3780 2000
471 57
2260 195
6 8
29 27
55 32 80
264 96 320
2.5 4.0 3.0
11.5 17.2 17.2
11.7 5.3
Prior (
Analyses
5-300
0-300
4-138
4-12
(*) Gas samples at two times 1 hour each for these runs.
(T) Gas samples throughout the period of this run.
Values obtained previously at Miami County, Oceanside, and Norfolk.
-------
Sampling and Analysis Methods
HC1 samples were collected by two methods: (1) a standard!* EPA
sampling' train consisting of a probe, heated filter, and two
impingers each containing 100 ml of 5 weight percent NaOH;fl the
chloride, analysis was done by titration with silver nitrate after
acidifying the solution; and (2) a continuous chloride monitoring
system based on a chloride electrode (Orion). The sampling system
consisted of a probe, heated filter, and a bubbler containing
500 ml of acetate buffer solution (pH 5.5) in which the chloride
electrode was immersed. The electrode output was recorded on a
strip chart to provide continuous readout of the potential change
from a chloride sensitive half-cell in the solution. Changes in
HC1 concentration in the furnace gases could be detected readily
by this method end the actual concentration of HCl was determined
by comparison with standard chloride solutions. Some discrepancies
were noted between chloride concentrations found by the two methods.
These differences are believed to result from relative absorption
efficiencies in the two systems.
Sulfur oxides were collected using a standard EPA train consisting
of a probe, heated filter and three impingers. The first' impinger
contained 100 ml of 80 percent isopropanal to absorb SOo. The
other two impingers each contained 100 ml of 3 percent H202 to
absorb S02« Both the SC>2 and the 863 were then determined by
titrati'on of the respective absorption solutions with barium
perchlorate, using thorin as the indicator.
Nitrogen oxides were collected by taking a grab sample in a 3-liter
evacuated flask which contained 25 ml of 3 percent H£02 acidified
with H2SOA. The analyses were made by the standard phenbl disulfonic
acid method.
Combustion-gases (carbon dioxide, oxygen, and carbon monoxide) were
determined by standard Orsat methods and Fyrite analyzer-s. The
water content was measured by condensing the moisture and determining
its volume.
Results of Gas Analysis
The data presented in Table 4 illustrate large hourly an well as
daily variations in nearly all of the components of the incinerator
gas. Although the average HCl contents from the two samplings
of the gas for Runs 17 and 18 and the amount obtained in the all-day
collection of Run 19/20 are higher than for Run 16, they do not
appear to be in proportion to the amount of PVC added to the refuse.
On the other hand, the chloride content in the deposits on the
corrosion probes do vary with the amount of the PVC addition to the
refuse (see Figures 12 and 13). Thus, it would appear that a
26
-------
large fraction of the HCl evolved from the PVC combines with the
metallic elements or their oxides to form inorganic chlorides rather
than remaining as HCl gas. Thus, the HCl content of the incinerator
gas is believed to be affected by other compositional variables in
the refuse.
A summary of the results of the two methods of collecting gas for
HCl analyses are presented in Table 5. It is evident here that
Method 1, the dual impinger alkaline solution, was a mere efficient
collector than Method 2, which consists of a single bubbler with an
acid solution. However, the latter gave an immediate response and
could be used to determine when the gases from the combustion of
PVC had reached the corrosion probe area of the incinerator. The
result of Method l. (average 860 ppm) for the 2 percent PVC addition
(Run 17) compares very well with the 816 ppm obtained by Kaiser(3)
when 2 percent PVC was added to the Babylon Incinerator. The
higher HCl content in Run 18 when only 1 percent PVC was added is
undoubtedly caused by other compositional variables in the refuse.
The other constituents of the gas environment, obtained on this
program, are consistent with previous analyses of furnace gas
composition at Miami County, Oceanside, and Norfolk incinerators.
DISCUSSION OF RESULTS
The results of corrosion rate, scale, and deposit studies illustrate
that the PVC content of refuse will have a marked effect on the life
of materials for heat recovery systems in municipal incinerators.
The corrosion of low alloy steels A106 and Til may be divided into
three temperature zones: < 400 F, 400-800 F, and 800-1200 F. In
the low-temperature zone, the corrosion rate decreases rapidly with
decreasing metal temperature while from 400-800 F, the corrosion
rate is essentially constant. Between 800 F and 1200 F, the corrosion
rate increases rapidly with metal temperature. The behavior in the
800-1200 F zone is very similar to that observed(6) with high (0.75%)
chloride coal having relatively low (0.970) sulfur, but the corrosion
rate of 0.07 mil/hr in 100 hour exposures at 1200 F when coal was
burned is approximately 1/5 that observed in an incinerator with
normal refuse and 1/10 that with a 2 percent PVC addition. This
difference, however, may be attributed in part to the furnace gas
temperature at the location of the specimens. As shown in the
present study, an increase of 150 F (1400 F to 1550 F) in gas
temperature can increase the corrosion rate of specimens maintained
at 1100 F, by internal cooling, by a factor of as much as 6X. If,
however, the specimens are held to below 800 F, the higher furnace
gas temperature does not affect the corrosion rate.
Although there is a significant (1.5-2X) increase in corrosion rate
at all temperatures by the addition of 0.5 percent PVC to the normal
refuse, the addition of larger (1% and 2%) amounts of PVC does not
continue to incraase the corrosion rates appreciably, when the
27
-------
TABLE 5. SUMMARY OF RESULTS ON HCl CONTENT OF GAS
HC1 Content, ppm
(*)
Method v '
1
2
3
4
Probe Run Numbers
16 17
370 860
210 801
343
364
18
904
448
375
392
1-9/20
583
71
98
104
(*) Methods 1 and 2 are described previously
while Method 3 involved titration of the
solution from 2 and Method 4 was X-Ray
fluorescence analysis of the solution
from 2.
28
-------
metal temperatures are held in the range of 400 to 800 F. However,
the corrosion rates of A106 and Til steels increase linearly with
PVC content for higher (1100 F) metal temperatures. Alexander(?)
observed the air oxidation of Type A213-T22 steel at 1100 F to
increase a factor of f^-f 2X by coating the steel with a 10 percent
NaCl"Na2SO^ mixture while in the present study, increasing the
PVC content of the refuse by 2 percent increased the chloride
content of the deposit to 4 percent and the corrosion rate of Til
at 1100 F by a factor of 2.5X from 0.22 to 0.55 mil/hr. While
the oxidation rates observed by Alexander are 1/100 those obtained
in the present work, it is important to note that her specimens
were exposed to air rather than to incinerator gases.
It appears that the corrosion mechanism is different above 800 F
from that below 800 F. This is confirmed by the composition of the
scale layer next to the metal in that below »UO F the layer is
composed of ferrous chloride while above 800 F, there is a mixture
of ferrous sulfide with pools of ferrous chloride on the metal
substrate. The initial increase in corrosion rate of the low-alloy
steels with a small increase in PVC content of the refuse for
temperatures below 800 F is attributed to the development of a
continuous layer of ferrous chloride on the metal surface. At
higher temperatures, however, the role of PVC in the refuse is
one of maintaining sufficient chloride ion to destroy the oxide
films and thus allow sulfide corrosion to occur, even though the
sulfur content of refuse is low compared to that of coal. It is
evident from analyses of the deposit that the amount of chloride
retained on the corrosion probes decreases rapidly with increasing
metal temperature. Perhaps volatization of ferrous chloride is
also contributing to the high corrosion rates in the upper
temperature zone of the corrosion probe. This would account for
the drastic corrosion observed by Heimburg(^) during the burning
of PVC.
Unlike the low-alloy steels, the corrosion rates of the stainless
steel were essentially unaffected by specimen temperature (800-1200 F)
or by the PVC content of the refuse. However, the corrosion rates
for the stainless steels were increased considerably (2X-6X) by an
increase of 150 F in gas temperature. This increase in corrosion
rate with gas temperature was observed for low-alloy-steel specimen
also when the metal temperature exceeded 800 F, but was not observed
at lower (400-800 F) metal temperatures. The lack of an effect of
PVC content in the refuse on the corrosion rates of the stainless
steels is undoubtedly due to the type of oxide formed and the low
retention of chloride in the deposits in the high temperature zone
on the probe. Alternatively, the base chloride content of refuse
may be sufficient to reach an equilibrium in corrosion rate versus
PVC content as was observed in the low temperature zone for the low-
alloy steel. Alexander^) found the oxidation rate of 316 stainless
steel coated with a 10 percent NaCl-NaSO^ mixture to be nearly the
same as that with 20 percent NaCl-Na2S04 mixture while with lower
amounts of NaCl the oxidation rates increased linearly with NaCl
content. The results presented in this study would indicate that the
29
-------
corrosion rates of the stainless steels, in the incinerator gases,
do not increase with chloride content of the deposit above about
1-2 percent chloride which is obtained with normal refuse. The
corrosion rate of 0.045 mil/hr for 316 stainless steel at 800-1200 F
in 1400 F incinerator gas is about SOX that obtained by;Alexnader
for the oxidation rate of 316 stainless steel coated with a 2 percent
NaCl-Na2S04 mixture. This large difference is undoubtedly attributable
to differences in gas composition between air and the incinerator gas
as previous laboratory studies at Battelle^ ' have shown that the
combination of mixed gases and mixed (17» NaCl-Na2SC>4) salts do produce
an attack of stainless steels at 1000 F nearly equivalent to that
observed in the present incinerator study. In the above mentioned
Battelle investigation, it was shown that both the mixed gas and
mixed salt environments were essential to approach the magnitude of
corrosion produced in an incinerator. Perhaps if gas temperatures
had been increased to that (1400 F) of the incinerator while
maintaining specijtien temperature, the corrosion rates would have been
the same. In the case of the stainless steels, however, the major
mode -of attack is intergranular which has been shown(^) to be
accelerated by the presence of chloride ions, particularly at
temperatures above 1000 F.
CONCLUSIONS
(1) The corrosion rates of the carbon and low-alloy steels
'increase sharply for small (0.5%) additions of PVC to
"municipal refuse.
(2) Larger (1% and 2%) additions of PVC does not affect the
.corrosion rate of low-alloy steels at temperatures below 900 F
,but increases the corrosion rate in direct proportion to the
amount added for metal temperature above 900 F.
(3) The general corrosion rates for the stainless steels are not
affected by either PVC content of the refuse or metal
temperature up to 1200 F.
(4) The corrosion resistance of the stainless alloys -to the
fireside environment of municipal incinerators decreases in
the order: 310, 446, 825, 316, and 321.
(5) All of the stainless alloys suffer grain boundary attack and
. are susceptible to stress corrosion cracking with the possible
exception of 446.
(6) Both the low-and high-alloy steels suffer accelerated (2X to 6X)
corrosion rates in the 900-1200 F metal temperature range when
the gas temperature is increased from 1400 to 1550 F.
30
-------
(7) The mechanism of attack at metal temperatures below 800 F is
essentially a chloride reaction with the substrate metal
while at metal temperatures above 800 F, the attack is
predominantly a chloride assisted sulfide reaction.
(8) The chloride content of the deposit increased with PVC
content of the refuse, but decreases with increasing specimen
temperature and with increasing furnace gas temperature.
(9) The sulfur content of deposits is greater on specimens from .
the high-temperature zone. Furnace gas temperature has no
appreciable affect on the sulfur content of the deposit.
(10) The chlorine (HC1) content of the furnace gas is increased
by PVC addition to the refuse, but is not proportional to
the amount added.
(11) Based upon the above conclusions, it is further concluded
that none of the alloys investigated would provide satisfactory
service in fireside heat recovery systems for municipal
incinerators.
ACKNOWLEDGEMENTS
The assistance of Mr. R. C. Thurnau and Mr. R. Loebker, of the Solid
Waste Research Division, Environmental Protection Agency, in the
analysis of types and composition of refuse and their encouragement
in the conduct of these corrosion studies were very valuable throughout
the investigation. In addition, the cooperation of Mr. R. Karnehm,
Superintendent of the Miami County Incinerator at Troy, Ohio, in
providing a facility and record of refuse received was a considerable
asset to the program. Battelle personnel contributing special
assistance to the program were Mr. E. White, Mr. W. Stiegelmeyer,
and Mr. J. Faught.
31
-------
REFERENCES
(1) Heimburg, R. V7. Environmental Effects of Incineration of
Plastics. AIChE Sixty-Eighth National Meeting. Houston,
Texas. February 28, 1971.
(2) Riemer, H., and Roesi, T. Mull and Abfallo.
March 21-24, 1970,
(3) Kaiser, E. R., and Carotti, A. A. Municipal Incineration of
Refuse with 2 Percent and 4 Percent Additions of Plastics:
Polyethylene, Polyethyrune, Polyurethane, and Polyvinyl
.Chloride. Report to Society of Plastic Industry.
June 30, 1971.
(4) Schonborn, H. H. The Situation of Waste in Germany.
Plastic EFTA International Symposium, Oslo, Norway.
May 26, 1970.
(5) Fessler, K., Leib, H., and Spohn. Mitteilungn der. VGB,
48(2):130, 1968.
(6) Johnson, H. R., and Littler, D. J. The Mechanism of
Corrosion by Fuel Impurities. Butterworths, London.
523, 1963.
(7) Johnson, H, R., and Littler, D. J. Ibid. 575
(P. Alexander).
(8) Miller, P. D., et al. Fireside Metal Wastage in Municipal
Incinerators. EPA Report SW 72-3-3. Research Grant
EP-00325, 1972.
(9) Pickering, H. W. Stress Corrosion of Austenitic Stainless
Steels by Hot Salts. Physical Metallurgy of Stress
Corrosion Fracture, 4. Interscience Publications,
/T* New York, 1959.
32
-------
APPENDIX
ANALYSES OF TYPES AND COMPOSITION OF REFUSE
The hetrogeneous nature of the refuse burned at the Miami County
incinerator results from the sources which includes a number of
industries plus municipal refuse from surrounding communities.
The ratio of municipal to industrial refuse received for the
week of June 26, 1972, was 2.08 with a total 878 tons (593 tons
municipal and 235 tons industrial). As the amount of each received
differs daily, the M/I ratio varied from 0.65 to 4.42. Wednesday,
June 28, 1972, had nearly the maximum in the amount (187 tons)
received and the M/l ratio was 2.03. Based upon this inspection
of the records at the Miami County incinerator, Wednesday was
selected as the day of the week to conduct corrosion experiments
to evaluate the effect of PVC on the corrosion of water wall
materials.
The program consisted of four runs with five corrosion probes and
exposure times of 10 hours each. The refuse received on the days
of these runs is tabulated below:
Run Weight of Refuse, tons
No. Day Municipal Industrial Total Ratio
16 8/9/72 117.5 53.5 171.0 2.10
17 9/13/72 82.5 57.0 189.5 1.45
18 10/25/72 86.5 62.0 148.5 1.40
19/20 12/13/72 110.0 40.5 150.5 2.72
As the incinerator operates on a 24-hour-a-day basis, approximately
10/24 of the refuse received was burned during the 10-hour
exposure period. No PVC was added during Run 16 to provide baseline
corrosion rate data for these short time experiments. For Runs 17,
18, and 19/20, PVC was added to each grapple load fed into the
incinerator. Approximately 160 grapple loads were fed into the
incinerator in the 10-hour exposure periods. As a grapple load
weighs ^ 1000 Ibs the amounts of PVC added were 20 Ibj (2%),
10 Ibs (1%), and 5 Ibs (1/2%) per grapple load for the Runs 17,
18, and 19/20, respectively.
33
-------
In addition to the above catorgization of refuse, Mr. R. Thurnau
and Mr. R. Loebker with the assistance of various Battelle staff
made a hand sort of a grab samples taken on each of the above
dates. These grr.b samples were taken at approximately 10:00 AM
from the refuse pit by the crane operator and placed on the floor
adjacent to the pit. The refuse was sorted into 12 categories,
weighed and percentages calculated as shown in Table A-l. The
total weight of the grab samples ranged from 303 to 785. Ibs (0.2-0_57o
of the refuse burned in 10 hours). A small but proportional amount
of each category was taken to the Solid Waste Research Laboratory
in Cincinnati for analysis. The categories were remixed and ground
to A.Q-and 0.5-mm size particles for analysis of the following
properties:
(1) Moisture content.
(2) Ash content.
(a) Composition of ash.
(3) Volatiles content.
(a) Composition of volatiles.
(4) Heat content.
The results of these analyses, as presented in Table A-2, were
supplied by Mr. Loebker except for the element analyses on the ash
which were made at Battelle by optical emission spectroscopy (OES).
The amounts of the various elements found in the ash are very similar
in alkali metal content to that found in the deposit, see Table 2
of this report. However, the deposits had significantly higher
concentrations of iron, zinc, and lead and considerably more chloride
and sulfate than the ash. This suggests that the cold wall of the
corrosion probe tends to preferentially retain some materials.
The higher iron content in the deposit is probably due to oxidation
of the probe.
-------
TABLE A-l. TYPE OF REFUSE IN GRAB SAMPLE (Percent)
U)
Category
Cloth
Plastic
Food
Yard and
Garden
Wood
Paper
Corrugated
Paper Boxes
Newspaper
Metal
Glass
Magazines
Fines
Total Weight
of Grab, in
Ibs
Run
Aug.
Wet
2.07
3.06
17.78
10.00
5.29
20.58
13.82
2.55
12.87
8.60
0.76
2.61
784.7
16
9, 1972
/o n«U
32.81
28.35
54.56
62.53
7.60
36.65
14.61
15.18
9.42
0.22
10.30
43.07
- 29.94
Run
Sept. 13
Wet .
2.53
5.57 .
3.08
0.04 ,
2.76 -
62.05 .,-;
5.79
2.30
6.19
4.16
1.56
4.98
642.25
17
, 1972
% H20
-------
TABLE A-2. ANALYSIS OF REMIXED AND GROUND'REFUSE FROM GRAB SAMPLE
Analysis
As Indicated
7, Moisture
7, Ash
7» Ash
7, Si in Ash
7, Ca in Ash
70 Na in Ash
% K in Ash
% Al in Ash
% Mg in Ash
7» Fe in Ash
7o Ti in Ash
70 Pb in Ash
7> Zn in Ash
7» Cu in Ash
7» Mn in Ash
7, Ba in Ash
7, Cr in Ash
7, Zr in Ash
7, Sn in Ash
% Ni in Ash
7o Mo in Ash
7» Cl in Ash
7» Loss on
ignition
7o C in dry
sample
% H
7, N
7» Cl
7, S
7»00
Btu/lb
Run 16
4 mm 0.5mm
29.94
15.10 16/37
16.2
10-20
5-10
2-4
2-4
3-6
1-2
3-5
2.0
0.3
0.4
0.2
0.2
0.3
0.04
0.02
0.02
0.005
0.005
0.011 0.006
84.90 83.63
42.92 41.82
5.60 5.46
0.57 0.78
0.54 0.44
0.20
35 o 1 34 . 9
7691 7548
Run 17
4 mm 0 . 5mm
26.81
10.24 11.69
12.1
10-20
5-10
5-10
2-4
5-10
2-4
3-5
3.0
0.4
0.2
0.2
0.1
0.1
0.04
0.04
0.02
0.02
0.005
0.020 0.018
89.76 88.31
45.06 42.94
6.13 5.76
0.54 0.56
0.79 0.70
0.20
37.0 38.2
8083 7742
Run 18
4 mm 0.5 mm
36.59
7.62
10.0
10-20
10-20
3-6
3-6
3-6
2-4
1-2
1.0
0.2
0.1
0.2
0.1
0.3
0.04
0.01
0.01
0.005
0.005
92.38
48.19
6.4,7
0.91 .
0.32 0.36
0.20
36.28
8886
Run 19/20
4 mm 0 . 5 mm
23.59
5.86 6.05
6V2
10-20
5-10
2-6
3-6
10-20
1-2
1-2
1-2
0.4
0.3
0.2
0.05
0.1
0.07
0.01
0.01
0.005
0.007
94.14 93.95
47.24 45.93
6.37 6.14
G'.75<*J
0.37 0.38
0.20
39.21 40.55
8523 8164
Remarks {f .- . /'•'
r1
L-0-100 C
LOI (960 C)
LOI (500 C)
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
OES
Solid Waste Laboratory
Solid Waste Laboratory
Solid Waste Laboratory
Solid Waste Laboratory
Solid Waste Laboratory
Solid Waste Laboratory ,
Based upon other refuse^ '
By difference
(*) Estimated on basis of average for similar samples.
(•-) Twenty-one other samples gave S contents of 0.1.0 to 0.37 percent,
averaging 0.19 percent.
-------