WATER POLLUTION CONTROL RESEARCH SERIES • 12060 EDZ 08/71
Pilot Plant Installation
for Fungal Treatment
of Vegetable Canning Wastes
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH. SERIES
The Water Pollution Control Research. Series describes the
results and progress In the control and abatement of pollu-
tion of our Nation's waters. They provide a central source
of information on the research, development, and demon-
stration activities of the Environmental Protection Agency
through inhouse research and grants and contracts with
Federal, State, and local agencies, research institutions,
and industrial organizations.
Inquiries pertaining to the Water Pollution Control Research
Reports should be directed to the Head, Publications Branch,
Research Information Division, R&M, Environmental Protection
Agency, Washington, D.C. 20460.
-------
Pilot Plant Installation for Fungal Treatment
of Vegetable Canning Wastes
THE GREEN GIANT COMPANY
LeSuer, Minnesota 56058
for the
ENVIRONMENTAL PROTECTION AGENCY
Grant No. 12060 EDZ
August 1971
For sale by the Superintendent of Documents, t) .8. Government Printing Office, Washington, D.0.30403 - Price $1
-------
EPA Review Notice
This report has been reviewed by EPA, and
approved for publication. Approval does
not signify that the contents necessarily
reflect the views and policies of the
Environmental Protection Agency, nor does
the mention of trade names or commercial
products constitute endorsement or recommenda-
tion for use.
ii
-------
ABSTRACT
The use of the imperfect fungus, TrLchoderma viride, to treat corn and
pea canning wastes has been tested in continuous fermentation systems
at the 10,000-gallon scale. Both a pool unit with a 2-hp floating
aerator and an oxidation ditch with a 3-ft rotary brush aerator were
tested. The pH was controlled to approximately 3.7 and ammonium
nitrogen and inorganic phosphate were added. The average residence
time was about 20 hours. An aerated lagoon was also operated to
compare with the two fungal systems.
In the fungal systems, about 96-percent removal of 5-day biochemical
oxygen demand (BODs), 88-percent removal of chemical oxygen demand (COD)
and 93-percent removal of total organic carbon (TOG) was achieved on
corn canning wastes. Performance on pea canning wastes was about
95-percent BODs removal, 81-percent COD removal, and 87-percent TOG
removal. Essentially zero levels of ammonia nitrogen and inorganic
phosphate could be attained in the effluent stream. Organic phosphate
levels were decreased by 80 percent.
Mycelium yields were equivalent, on a dry-weight basis, to about 50
percent of the BODs of the feed. The nitrogen content of the dry
mycelium was equivalent to about 50-percent protein. The most promising
fungal recovery system was a vibrating screen for bulk harvest and a
sand filter to remove materials passing the vibratory screen.
Costs are estimated at 4.9 cents per pound of BODs. Sale of mycelium
as feed could decrease this to 3.1 cents per pound of BODs. Operation
on a year-around basis with sale of the mycelium as a feed (assuming
simplified filtration requirements) would decrease costs to about 1.1
cents per pound of BODs. Direct feeding of the mycelium without drying
could further reduce the net cost significantly, to about 0.8 cents per
pound of
iii
-------
CONTENTS
Page
I CONCLUSIONS. ... ................... 1
II RECOMMENDATIONS ..................... 5
III INTRODUCTION ...................... 7
IV METHODS ......................... 9
Physical Facilities ................. 9
Sampling Procedures .... ............. 12
Analytical ...................... 12
Inoculation .................... . 13
Harvesting. .... ............. .... 14
Operating Calender ...... . . .......... 14
V RESULTS ......................... 15
Corn Waste. . . ...... .... ......... 15
Operating Parameters ......... ...... 15
Removal of Organic Matter ............. 23
Yield of Mycelium. .... ............ 23
Oxygen Consumption . ............... 23
Use of Nitrogen and Phosphate ........... 29
Microbial Pattern ................. 29
Recovery of Mycelium ....... ... ..... 32
Drying ...................... 33
Other Corn Waste Studies ............... 34
First Season Pilot-Plant Operation ........ 34
Study of Lagooned Wastes ............. 34
Temperature Effects ................ 38
Pea Wastes ...................... 40
Operating Conditions ............... 40
Removal of Organic Matter ............. 43
Mycelium Recovery ......... . ....... 46
-------
CONTENTS (Continued)
Section Page
Oxygen Consumption 47
Microbial Pattern 47
Corn Silage Wastes 49
Mechanical Performance 50
VI DISCUSSION 53
General Effectiveness 53
Organic Compound Removal .... 53
Nitrogen and Phosphate Removal 55
Acid Usage 55
Retention Times 56
Oxygen Requirements 56
Temperature Effects 57
Microbial Stability 57
Harvesting 57
Inoculation 58
Mechanical Performance 58
Comparisons with an Extended Aeration Process .... 59
Description of the Facilities 59
Evaluation of Performance 62
VII ECONOMIC ESTIMATES 69
VIII ACKNOWLEDGEMENTS 73
IX REFERENCES 75
X LIST OF PATENTS AND PUBLICATIONS 77
vi
-------
FIGURES
No. Page
1 Schematic Diagram of the Pilot Plant Flow System. ... 10
2 Aerated Ditch Used for Treatment of Corn Canning
Waste 11
3 Aerated Pool Used for Treatment of Corn Canning,
Pea Canning and Silage Wastes 11
4 Temperature of Corn Waste Treatment System
Measured at 2 p.m. Each Day 16
5 Retention Time of Corn Waste in the Ditch and in
the Pool 17
6 Total Loading of Ditch in Pounds of BOD5 or
COD per Day During Operation on Corn Canning Wastes . . 18
7 Total Loading of Pool in Pounds of BOD5 or
COD per Day During Operation on Corn Canning Wastes . . 19
8 Ammonium Sulfate and Sodium Dihydrogen Phosphate
Additions to Ditch in Pounds per 1000 Gallons of
Corn Feed 20
9 Ammonium Sulfate and Sodium Dihydrogen Phosphate
Additions to Pool in Pounds per 1000 Gallons of
Corn Feed 21
10 Sulfuric Acid Additions Required to Maintain pH
at About 3.7 During Operation on Corn Canning Wastes. . 22
11 COD of Feed and Effluent Streams of the Ditch
During Operation on Corn Canning Wastes 24
12 BOD5 of Feed and Effluent Streams of the Ditch
During Operation on Corn Canning Wastes 25
13 COD of Feed and Effluent Streams of Pool During
Operation on Corn Canning Wastes 26
14 BOD5 of Feed and Effluent Streams of Pool During
Operation on Corn Canning Wastes 27
vii
-------
FIGURES (Continued)
No. Page
15 Dry Weight of Solids Recovered from Effluent
Stream by Filtration During Operation on Corn
Canning Wastes 28
16 Dissolved Oxygen Concentration Each Day at 2 p.m.
During Operation on Corn Canning Wastes 28
17 Concentration of Phosphates in Feed and Effluent
of Ditch 30
IS Ammonium Ion Concentration of Effluent from Ditch
During Operation on Corn Canning Wastes 31
19 Performance of Aerated Pool in Continuous Treatment
of Corn Waste by T. viride 35
20 Continuous Laboratory Digestion by T. virn.de of
Corn Wastes Drawn Directly from Plant Effluent 36
21 Continuous Laboratory Digestion by T. viride of
Corn Wastes Drawn from the Receiving Lagoon 37
22 Performance of Laboratory Treatment of Lagooned
Wastes at Two Temperatures 39
23 Feed Identity and Retention Time in Pool 40
24 (NHi+)2SOi+ and NaH2POi+ Additions to Pool in
Pounds per 1000 Gallons of Feed 41
25 Sulfuric Acid Additions Required to Maintain pH
at About 3.7 42
26 Temperature of Pool Contents at 11 a.m. Each Day. ... 42
27 Total Loading of Pool in Pounds BOD5 or COD per day . . 43
28 COD of Feed and Effluent Streams 44
29 BOD5 of Feed and Effluent Streams 45
30 Dry Weight in Grams per Liter of Solids Filtered
from Effluent Stream 47
31 Dissolved Oxygen Concentration in Pool at 11 a.m.
Each Day 48
viii
-------
FIGURES (Continued)
No. Page
32 Flow Velocities, Dissolved Oxygen and COD Values
at Selected Sampling Points in the Aerated Pool .... 51
33 Pilot Lagoon for Extended Aeration at Glencoe,
Minnesota 61
34 BOD5 Sample Results for Pond 64
35 Dissolved Oxygen, Temperature, and Removal
Efficiency in Pond 65
36 Effect of Temperature and Detention Time on
Removal Efficiency in Pond 66
ix
-------
TABLES
No. Page
1 Growth of Inoculum 13
2 Characteristics of Composite Sample of
Corn Waste 16
3 Performance of Sweco Vibratory Filter on
Corn Waste Effluents 33
4 Analysis of Feed and Effluent Pea Wastes 46
5 General Efficiency of Fimgi Imperfeoti Process 53
6 Characteristics of Composite Samples of Corn
Waste Feed and Effluent Collected in Mid-Season .... 54
7 Cost Estimates on Corn Processing 69
-------
SECTION I
CONCLUSIONS
1. The use of Fungi Imperfeati in treating corn and pea canning wastes
was investigated on a pilot-plant scale. Pilot units used were a plastic
swimming pool holding 10,000 gallons of waste and an aerated ditch
holding 11,000 gallons. The pool was aerated with a 2-hp floating
aerator and the ditch with a 3-foot cage rotor. The installations were
made at the Green Giant plant at Glencoe, Minnesota. The pool was
operated in the latter part of the 1969 season, and through the 3-1/2
months of the 1970 canning season. The ditch was operated during the
corn canning season of 1970.
2. Both units were operated as continuous fermentations with average
residence times of 18 to 22 hours, although longer residence times were
used experimentally during part of the season. Sulfuric acid was added
automatically as required to maintain the pH at about 3.7. Ammonium
sulfate and sodium dihydrogen phosphate were added continuously at rates
believed needed for vigorous growth and production of a mycelium of high
protein content. Inoculation was with Triohoderma viride.
3. The rationale for the use of Fungi Imperfeoti was one of converting
dissolved and suspended organic matter into a mycelium that not only has
a high enough protein content to have value as an animal feed, but also
is large enough to be readily recovered by simple filtration or screening.
Laboratory studies had indicated that fungi could be maintained as the
dominant microorganisms if used as an inoculum and if the pH was kept in
the range of 3 to 4. Criteria of successful operation include lowering
of biochemical oxygen demand (8005) to low levels in the effluent stream,
low levels of nitrogen and phosphate in the effluent stream, ease of
harvest of the mycelium, high yields of mycelium, high protein levels in
the mycelium, economy of operation, and stability of microbial flora.
4. BOD5 removal from corn canning wastes was about 96 percent, to yield
an effluent stream of about 50 mg per liter BODs- COD removal was about
88 percent. Removal of total organic carbon (TOG) was about 93 percent.
These performances could be improved to decrease residues by about one
quarter by more complete clarification of the effluent. Performance on
pea canning wastes was about 95 percent BOD5 removal, 81 percent COD
removal, and 87 percent TOC removal.
5. On corn wastes, it was possible to produce an effluent stream with
lower levels of ammonia nitrogen and inorganic phosphate than could be
detected by the analytical methods. The detection limits were about
1 mg per liter of nitrogen and 0.2 mg per liter of phosphate. Levels of
organic phosphorous in the effluent stream were about 2.5 mg per liter.
Levels in the feed stream were about 12 mg per liter. Excellent nitrogen
and phosphate removals were attained only during part of the season
because of inadvertent over-feeding with these nutrients. Nitrogen and
phosphate removal were not investigated on pea wastes.
-------
6. Mycelium yields were judged to be satisfactory and were equivalent
on a dry-weight basis to about half the BODs level of the feed stream.
7. Harvest of mycelium in the pilot plant was considerably more
difficult than in the laboratory because a finer mycelium was obtained.
One reason for the finer mycelium may be the more violent mechanical
action of the aerators, although it has not been proven that this is the
case. Another reason has been the appearance of fungal species other
than T. vi-ride with very small fine floes. There was some evidence that
low temperatures contributed to a finer mycelium during part of the
season. The mycelium could not be efficiently recovered on a stationary
screen as had been possible in the laboratory. About two-thirds of the
mycelium could be recovered on a vibratory screen unit manufactured by
the Sweco Company at a flow rate of about 6 gallons per square foot of
filter area per minute. Several ways of recovering the material that
passed through the vibratory screen unit were identified, with the most
promising being the use of a sand bed filter. Twenty bed volumes of
material could be filtered at flow rates above 2 gallons per minute per
square foot of surface. The mycelium could be readily recovered by back-
washing. The effluent from the sand bed had a turbidity of about 50 by
the Jackson candle test . Once the mycelium had been obtained in relatively
concentrated form, it could be dewatered on a vacuum filter to yield a
cake of 20 percent solids content. Further drying could be accomplished
at temperatures up to 100°C to yield a product of light brown color and
little taste. Layers of drying fungal material had to be kept thin to
avoid surface hardening.
8. Protein levels in the mycelium were examined only by analysis for
total nitrogen. By this measure they were in the neighborhood of 50
percent .
9. The microbial flora remained fungal in type, but underwent shifts
away from dominance by Triahoderma vim.de. On corn wastes, the pool
remained T. wii*idef but the ditch changed to as high as 70 percent
Geotrichion during midseason. It later gradually returned to a greater
percentage of T. viride. It is speculated that oversupplying nitrogen
and phosphorus could have had a role in these shifts. Laboratory studies
strongly suggested that changes occurring in waste temporarily stored in
the receiving lagoon were deleterious to T. viride. The pea waste
fermentation quickly became dominated by a Fu8ariimt which matched
laboratory experience with this waste stream.
10. Several factors increased costs of operation over those estimated
from laboratory experience. One was higher oxygen requirements. The
oxygen requirement is now estimated at 0.7 pound per pound of BOD 5 re-
moved. There remains a degree of uncertainty in this estimate. Another
factor increasing costs was higher sulfuric acid requirements. This is
a function of water hardness, fermentation in the receiving lagoon, and
BOD 5 concentration. Avoiding }.ong residences in the receiving lagoon
and a more concentrated waste stream would lower this cost per unit of
-------
BODs significantly. A third, and very important factor, was the in-
creased investment in harvesting equipment needed to accommodate the
finer mycelium. Our present estimates of treatment costs are 4.9 cents
per pound of BOD5. Sale of the product might lower the costs to 3.1 cents
per pound of BODs. *n making these calculations, a 3-month-per-year
operating season has been assumed. Operating on a year-around basis
would spread investment costs over a greater waste volume and lower
estimates of treatment costs with sale of the recovered solids, to about
1.1 cents per pound of 8005. Elimination of the drying step could
reduce the net cost to about 0.8 cents per pound of BODs. ^ ^s obvious
that further economies could be achieved by use of a more concentrated
waste stream, or use of softer water. Some economy could be achieved
by use of shorter retention times, and we have no evidence that we
were approaching the lower limits of retention time except when
temperatures were below 15°C.
11. It is concluded that the Fungi. Imperfeeti system has performed
credibly on a pilot-plant scale in yielding a purified effluent stream,
but that costs of operation are higher than desired. Several
possibilities of lowering costs are visualized and several applications
can be recognized where costs would be lower, even without substantative
improvements in the system.
12. An aerated lagoon was constructed late in the corn canning season
and was operated along with the fungal systems. Although the lagoon
covered only a partial season, performance efficiency based on BODs
removal per horsepower per day showed 92 percent. This corresponded
to 95 percent BODs removed in the fungal systems.
-------
SECTION II
RECOMMENDATIONS
This study had as its objectives the determination of the feasibility of
using selected imperfect fungal strains for the degradation of vegetable
canning wastes, harvesting the fungal solids, and employing the fungus
as the protein component in animal feed formulations.
With the establishment of the feasibility of these objectives, both in
previous laboratory studies and in the studies reported here, recommenda-
tions are made for further development of this important process.
1. A larger-scale (50,000 gallon) pilot plant should be built to
operate for at least ten months of the year.
2. Studies should be carried out on start-up procedures employing
laboratory and pilot development of dried fungal inoculum.
This material might be held in cold storage as spores or
mycelium dried in vegetable waste substrate.
3. Storage survival studies are needed to ensure the viability of
the dried inoculum for process start-up.
4. Necessity for, and means of, smoothing out the variations in
composition of a mixed vegetable waste stream, in salt con-
centration, and in flow should be explored.
5. The effects of the methods of harvesting and drying on
mycelium quality must be determined by studies to permit use
of the final harvested solids as the protein component in
animal feeds. Such a use would allow recycling of the
organic waste materials and convert carbohydrate-containing
wastes to useful materials.
6. Applications to waste from other food or nonfood processes
should be sought. These might include wet corn milling,
meat processing, potato processing, tomato canning, paper
manufacturing, and others.
7. Studies to determine the feasibility of harvesting large
quantities of crude enzymes; e,g.} amylases, celluloses,
proteases, and lipases, should be conducted.
-------
SECTION III
INTRODUCTION
Investigation of a possible role for organisms of the Fungi Imperfeoti
class for treating food processing wastes was undertaken on the hypothesis
that the fungi would be efficient in converting dissolved and suspended
organic matter into a mycelium which could be readily recovered and which
would have value as an animal feed. Laboratory studies1 using wastes
from corn canning and from soy protein isolation gave further encourage-
ment that these hopes could be realized. .Over 98-percent removal of
BODs was achieved in a 20-hour retention time. The equivalent of over
50 percent of the BOD5 was recovered as mycelium. The mycelium had a
favorable amino acid pattern and gave good growth of rats in limited
feeding trials.
The present investigations were undertaken to determine if the promise
shown in the laboratory studies could be corroborated in a field installa-
tion and with commercial equipment. Pilot-plant operation has extended
over two operating seasons, but had a very late start the first season.
Pea wastes were included in the second season of operation.
An aerated lagoon was operated in conjunction with the fungal pilot
plants for comparative purposes.
Funding for the investigation was supplied by EPA, the Green Giant
Company, the Wisconsin Canners and Freezers Association, the Minnesota
Canners Association, and the National Canners Association.
-------
SECTION IV
METHODS
Physical Facilities
Pilot-plant installations were made at the Green Giant canning plant at
Glencoe, Minnesota. Two types of aeration systems were used: a floating
aerator installed in a plastic swimming pool, and a cage rotor installed
in a circular ditch. The floating aerator was manufactured by Richards
of Rockford, and was equipped with a 2-hp motor. During part of the first
season the floating aerator was equipped with an impeller intended to
reduce its output to one hp. An impeller which reduced output to the
1.5-hp level was used while operating on pea and silage wastes in the
second season. Aeration of the ditch was accomplished with a 3-ft cage
rotor aerator of 5-hp capacity, manufactured by Lakeside Engineering
Company.
The pool was 23.6 feet in diameter and was filled to a depth of 3.2
feet. At this depth, it held approximately 10,000 gallons. The ditch
was sized to hold 11,000 gallons. It was V-shaped in cross section and
was a circle of 100-ft center-line circumference. The liquid level was
3.1 feet, and the V was about 8 feet across, at the surface of the liquid.
The ditch was lined with rubber sheeting. In both installations, feed
rate of plant effluents was regulated by a constant head tank discharged
through calibrated orifices. Exit was by overflow. The feed intake for
the pilot facilities was from the first of a series of lagoons in which
the plant wastes are normally treated. The intake was initially 100
feet from the plant outlet, but later in the season it was moved to a
position within 10 feet of the outlet. The lagoon receiving the plant
waste held 13 million gallons and had an average retention time during
the operating season of 4 to 15 days.
Other units in the .physical plant were plastic storage tanks and
metering pumps to provide input of ammonium and phosphate salt solutions,
and plastic storage tanks and a recording pH meter which actuated a
metering pump for acid addition. The installation is indicated in Figure 1.
The ditch and pool units are pictured in Figures 2 and 3.
-------
Plant Waste
Stream
Figure 1. Schematic Diagram of the Pilot Plant Flow System
-------
Figure 2. Aerated Ditch Used for Treatment of
Corn Canning Waste
Figure 3. Aerated Pool Used for Treatment of
Corn Canning, Pea Canning and Silage Wastes
;;
-------
Sampling Procedures
A technician was at the pilot-plant site for eight hours each day (three
hours on Sunday) to make observations and to take samples. Samples of
feed and effluent from both the ditch and pool were taken at the same
time each day (11: a.m. during pea canning, 2:00 p.m. during corn
canning). The effluent samples were filtered through Whatman No. 4
filter paper and frozen until they could be transported to the laboratory
for analysis. Twice a week unfrozen, fresh, samples were also processed
in the laboratory. There was no difference in results between the frozen
and unfrozen samples taken on the same day. The filter papers had been
dried and weighed before use. They were sent to the laboratory to
permit determination of the amount of filterable material (mycelium) in
a known volume of the effluent. Samples of feed were also frozen for
transport to the laboratory without prior filtration. Since the feed
was drawn from a receiving lagoon rather than directly from the plant
outlet, it was believed to be adequately mixed so that single samples
were representative.
Readings of dissolved oxygen (DO) levels and temperature were taken with
a Yellow Springs Model 54 oxygen meter when the samples were drawn.
Routinely, the probe was placed two feet from the edge of the pool, at a
depth of 1.5 feet. In the ditch, it was placed just upstream of the
cage rotor at a depth of two feet. At intervals in the course of the
investigations, a recording dissolved-oxygen meter was available so that
readings could be made around the clock. The pH was constantly monitored
and recorded. The sulfuric acid storage tanks were calibrated so that
the amount of acid used each day could be estimated with fair accuracy.
Nutrient feed pumps were set to deliver desired amounts of nutrient
solution and were checked each day to make certain that the desired
amounts were actually being delivered. Similarly, the feed rates of
incoming waste were checked daily.
At least every third day samples of unfiltered effluent were drawn and
taken to the laboratory for microscopic examination.
Analytical
Chemical oxygen demand measurements were made as described in Standard
Methods for the Examination of Water and Waste Water.2 Measurements of
BOD5 were carried out according to standard methods.2 Nitrogen determina-
tions on the wastewater were made by the Conway micro-diffusion method.3
This method involves release of ammonia following addition of strong
alkali, and is useful only for the determination of ammonium ion.
However, it was suitable for determination of the amount of ammonium
sulfate needed to supply the nitrogen requirements of the fungi. The
nitrogen content of mycelium was determined by the micro-KJeldahl
method.4 Phosphates were determined by the method of Fiske and Subbarow,5
12
-------
applied with and without prior acid hydrolysis. Without acid hydrolysis,
the method measures only inorganic phosphorus; after acid hydrolysis, it
measures total phosphate as phosphorus.
Inoculation
Inoculation was, in each case, effected by use of a culture of Trichoderma
viride 1-23 grown in four 32-gallon plastic garbage cans. The cans were
equipped with plastic pipe inserts with small holes along their length
for introduction of air from a compressed air source. Each of the cans
was1 inoculated from a shake-flask culture which, in turn, had been
inoculated from a test tube slant culture. The medium used for growth
of the inoculum was ground corn plus appropriate amounts of ammonium
sulfate, sodium dihydrogen phosphate, and sulfuric acid. Successive
additions of this nutrient were made as fungal growth occurred.
Successive additions were made richer in total nutrient. In this way,
a final inoculum of 80 gallons containing 5.6 grams of mycelium (dry
weight) per liter was attained in four days. Steps in the growth cycle
of the inoculum for each can are shown in Table 1.
Table 1. Growth of Inoculum
Hours after
Inoculation
0
24
48
72
112
Additions
Volume
(liters)
20
20
20
20
20
Concentration
Solids
(g/liter)
1.1
2.3
10.7
19.1
—
Dry Weight
Fungal Mass*
(g/liter)
—
0.8
1.4
2.9
5.6
*To some degree the fungal mass included suspended solids from
the ground corn.
The pool and ditch each contained about 4500 gallons of canning waste at
the time of inoculation. Feed was either started very slowly or delayed
until the culture became established.
13
-------
One variant of this procedure was the use of equal amounts of ground
peas and corn to prepare inoculum for pea canning wastes.
Another variant used in the first season was to use corn canning wastes
as the medium for inoculum propagation; a continuous culture was simulated
by dipping out part of the culture at regular intervals during growth.
and replacing the volume removed with fresh corn wastes to which the
required inorganic nutrients had been added.
Harvesting
A Sweco, Inc. Model LC-18-C-333 filter unit, with a one-square-foot
filter surface, was used on-line during part of the season. Other
mycelium recovery techniques were tried on a laboratory scale during the
same period.,
Operating Calender
The pool unit was operated on corn canning wastes during the 1969
canning season (September 12 until October 13). The canning operation
ended on September 20, so subsequent operation depended entirely on
wastes held in the receiving lagoon. In 1970, the pool unit was
operated on pea canning wastes from June 22 until August 8. Silage
juice supplemented the peas wastes from July 26 until August 8. The
pool was operated on corn canning wastes from August 31 until
September 29. Two failures of pH control instruments rendered pool
operations ineffective between August 8 and August 31. The ditch unit
was operated in 1970 on corn canning wastes from August 9 to October 2.
Canning plant operation ceased on September 22.
14
-------
SECTION V
RESULTS
Corn Waste
The bulk of the detailed results on treatment of wastes was obtained
during the second season of operation. Except as otherwise noted, the
data reported are taken from the second season.
Operating Parameters
Variables which might affect the operation of the waste treatment system
are:
• Nature of the feed
• Organism used for inoculation
• pH control
• Temperature
• Retention time
• Inorganic nutrient addition
• Oxygen level
• Stirring
Some of these variables were controlled by the operator; some were not.
Some of the characteristics of the feed are indicated by measurements
made on a composite sample composed of equal aliquots from frozen and
thawed samples obtained from the heart of the operating season (from
August 19 to September 20). Results of the examination are shown in
Table 2.
The temperature pattern encountered during the operating season is shown
in Figure 4. Measurements were made at 2 p.m. each day. It should be
noted that the temperature of the ditch was characteristically about one
degree higher than that of the pool. This is thought to be caused by the
protection provided by its sunken position.
The effect of variations in retention time was investigated; the time was
also altered occasionally in response to other events. It was deliberately
maintained high (a low flow rate) immediately after inoculation. Also,
at day 21, the retention time in the ditch was reduced for two days
because a failure of the pH control equipment had allowed the pH to drop
15
-------
Table 2. Characteristics of Composite Sample
of Corn Waste
COD
BOD 5
TOC
Total solids
Filterable solids (suspended)
Volatile solids
Ash
pH
Ammonium nitrogen
Phosphate, as P
Acid to titrate to pH 3.7
30
o
o
£
3
1 20
a.
— Ditch
• - Pool
IQ
0
10
20 30
Day of Operation
40
50
60
Figure 4. Temperature of Corn Waste Treatment System Measured
at 2 p.m. Each Day
16
-------
to levels that impaired culture performance. The pool was operated at
longer retention times than desired during part of the season because of
partial failure of a sump pump. (The retention times are represented
in Figure 5.) The retention times, together with the COD or BODs levels,
governed the daily loading, which are shown in Figure 6 for the ditch
and in Figure 7 for the pool.
r eo
a
o
X
1
-60
c
0
g
» 40
£K
20
I
m
—
•
—
—
^^
i
)
1 ' 1 ' I ' 1 ' 1
Ditch
Pool
•
—
_
— i p-i 1 __.
j _•
I
i
1
^ 11 11
1 . 1 • 1 . 1 • 1 .
10 20 30 40 50 6C
Doy of Operation
Figure 5. Retention Time of Corn Waste in the Ditch
and in the Pool
17
-------
300
250
o
O
i.
S-200
o
o
o
o
in
0 150
CD
<*-
O
(A
100
50
I
10
20 30
Day of Operation
40
50
Figure 6. Total Loading of Ditch in Pounds
of BOD5 or COD per Day During
Operation on Corn Canning Wastes
18
-------
200
SJ50
o
o
a
o
in
§100
CD
-------
6.0
50
o
o
o
2 4.0
L.
0)
Q.
Q.
CVJ
3.0
O
J^ 0.4
^
x
NaH2P04
0.2
O
a.
y
I
I
10
20 30
Day of Operation
40
50
Figure 8. Ammonium Sulfate and Sodium Dihydrogen
Phosphate Additions to Ditch in Pounds
per 1000 Gallons of Corn Feed
20
-------
6.0
o
o
o
&4.0
I
I
Q.
CM
X
O
Z
£3.0
CO
I
5 0.4|-
NaH2P04
I
I
I
I
I.
\
\_
10 20 30 40 50
Day of Operation
Figure 9. Ammonium Sulfate and Sodium Dihydrogen
Phosphate Additions to Fool in Pounds
per 1000 Gallons of Feed
21
-------
The pH control was set at 3.5, and that pH was maintained within +0.1
unit, with few exceptions. A major exception in the ditch occurred on
the twenty-first day, when the acid pump failed to turn off and the pH
dropped to 1.8. The pH control failures occurred in the pool because
of a loose pH probe connection. The pH meter showed the pH to be in the
proper range when actually it was below 2.0. This destroyed operations
in the pool twice during the first 20 days of the corn canning season.
The amount of sulfuric acid required to maintain the pH is indicated in
Figure 10. The amount of acid averaged about four pounds per 1000
gallons of feed. Titrations of the well water used in the canning
operations showed an acid requirement of about 2.9 pounds per 1000
gallons to bring the pH to 3.5. Titrations of the plant waste directly
from the corn canning operations showed a similar requirement. Titrations
of plant waste drawn from the receiving lagoon, however, showed a require-
ment of six pounds of acid per 1000 gallons.
•»
§30
CD
O
O
O
;20
S
CO
CM
X
«•-
O
w
10
Ditch
Pool
10
20 30
Day of Operation
40
Figure 10. Sulfuric Acid Additions Required to
Maintain pH at About 3.7 During Operation
on Corn Canning Wastes
22
-------
Removal of Organic Matter
The performance of the ditch and pool in removing COD and BOD5 are shown
in Figures 11 through 14. These data were taken from Whatman Number 4
filtered effluent samples. COD removal during favorable periods of
operation of the ditch (for example, between Days 25 and 36) was about
89 percent. The high effluent COD on Day 21 followed failure of pH
control with a fall of pH to 1.8. The peak on Days 37 and 38 occurred
when the temperature in the ditch and pool dropped to below 10°C. BODs
removal was about 97 percent during periods of favorable operation.
BOD5 levels in the effluent vacillated between about 40 and 90, with a
mean of about 50, except during upsets caused by extremes of acidity or
temperatures. It is to be noted that the decreased COD removal caused
by extreme acidity on Day 21 is also reflected by a peak in effluent
BOD5. The increase in COD levels in the effluent during the period of
low temperature on Days 37 and 38 is not reflected by a corresponding
increase in effluent BODs. Effluent COD and BODs levels from the pool
show slightly higher values than those of the ditch; particularly COD
values.
COD values of the feed showed more variation from day to day than did
BOD5 values. COD and BOD5 values were initially close together, but by
Day 20, had separated appreciably. The ratio of BOD5 to COD values in
the feed from Day 20 to Day 50 averaged 0.55. The ratio of BODs to COD
values in the effluent averaged about 0.24, with extremes of 0.12 and 0.44,
Yield of Mycelium
The yield of mycelium as recovered from daily effluent samples on
Whatman No. 4 filter paper varied from about 0.55 gram per liter to 0.8
gram per liter. Values are shown graphically in Figure 15. In both the
ditch and the pool the dry weight of recovered mycelium averaged 50
percent of the weight of BOD5 removed. The amount of mycelium recovered
was relatively constant. The spikes on Days 6 and 9 for the ditch are
thought to be caused by imperfect sampling procedures.
Oxygen Consumption
The dissolved oxygen levels in the pool and ditch at a common time each
day are shown in Figure 16. In interpreting this figure, it must be
remembered that the pool was equipped with a 2-hp floating aerator unit
and the ditch, with a 5-hp cage rotor unit. There is no question that
the aerator in the ditch was more powerful than required; therefore,
little information on power requirements for aeration in the ditch was
obtained. Dissolved-oxygen levels in the pool approached zero for several
days in a row on two occasions, however, indicating full utilization of
the aerator capacity. The BOD5 removal during these periods might be
used to estimate aeration requirements. If one assumes the floating
23
-------
3500
Peak COD of Feed on
23rd day was 3904
mg per liter.
3000
2500
=2000
ff
o
o
o
1500
1000
500
I
I
10
20 30
Day of Operation
40
50
60
Figure 11. COD of Feed and Effluent Streams of the Ditch
During Operation on Corn Canning Wastes
24
-------
3000
I 'I • I
2500
2000
in
8 1500
m
1000
500
10
20 30
Day of Optration
40
50
60
Figure 12. BOD 5 of Feed and Effluent Streams of the
Ditch During Operation on Corn Canning
Wastes
25
-------
33UU
3000
2500
2000
(-^
1
§1500
u
1000
500
Q
A /\
/ V / V
/ \ / \
-A X \
A** A
Feed \
1
! A
i / \
\/ \
\ 17 N
\ \
\ *
I
I
\
- \
\
^ Effluent
i
i A /^,
I' V\j V
1 . 1 . 1 . 1 . 1
0 10 20 30 40 50 60
Day of Operation
Figure 13. COD of Feed and Effluent Streams of Pool
During Operation on Corn Canning Wastes
26
-------
2500
2000-
= 1500
o»
o
00
1000
500
V *
X
Effluent
10
20 30 40
Day of Operation
50
Figure 14. BOD5 of Feed and Effluent Streams of Pool
During Operation on Corn Canning Wastes
aerator can transfer two pounds of Q£ per hp-hour, the 2-hp unit will
transfer 96 pounds in one day. From Days 26 to 31 (a period when the
DO was near zero), 75 pounds of BOD5 were being removed per day; on
Days 40 to 44, when the DO was measured at 1.0 to 1,8 mg per liter, the
unit was handling 85 pounds of BOD5 a day. Using these periods as a
basis for calculation, one arrives at an estimate of 1.2 pounds of
dissolved oxygen per pound of BOD5 removed. During the first of the
two periods, the detention time was greater than 40 hours; during the
second, the time was about 30 hours. Both detention times were higher
than for best operation, and performance was probably considerably below
27
-------
I.S
UJ
•5 '0
I
i
0.5
2
(9
— Ditch
—Pool
I
10
20 30
Day of Operation
40
50
Figure 15. Dry Weight of Solids Recovered from
Effluent Streams by Filtration During
Operation on Corn Canning Wastes
12
Ditch
Pool
8
x
O
o 4
8 4
I
10
Figure 16.
20 30
Day of Operation
40
50
Dissolved Oxygen Concentration Each Day
at 2 p.m. During Operation on Corn Canning
Wastes
28
-------
optimum. A further question is raised when the performance from Days
35 to 38 is considered. Then the unit was handling about 120 pounds
of BOD5 a day with DO levels averaging 8.0. During this period, based
on the same oxygen transfer assumption, the oxygen use must have been
less than 0.8 pounds of dissolved oxygen per pound of BOD5 removed.
During this period the temperature dropped to a low of about 5°C
(see Figure 4), which would permit greater oxygen transfer, but would
not account for the lower oxygen requirement. Reference of Figure 14
shows that BODs levels in the effluent were satisfactorily low during
this interval.
Probably a better approach to estimation of oxygen use was to measure
the rate of change in DO in samples removed from the ditch. These
measurements were made in a flask that allowed no air space above its
contents, and stirring was provided with a magnetic stirrer. The flask
was fed fresh feed from a buret at a rate that would give the same
retention time in the flask as in the ditch. The assumption was that
the oxygen use in the flask sample continued at a rate that allowed
removal of 95 pounds of BOD5 per day — the rate of removal being
accomplished in the ditch at the time of these measurements. Oxygen
use was 3 mg per liter per minute in several measurements. For a ditch
volume of 11,000 gallons, the oxygen consumed then would have been 65
pounds per day, or 0.7 pound per pound of BODs removed.
Use of Nitrogen and Phosphate
Levels of phosphate and ammonium ion in the effluent and feed for the
ditch are given in Figures 17 and 18, respectively. Levels of phosphate
in the effluent were lower than in the feed in spite of phosphate
additions to the fermentation. During one interval, between Days 20 and
23, phosphate levels were below 0.1 mg per liter. Ammonium nitrogen
levels were also too low to measure at intervals during the operations
(Days 12-14 and Day 48). Actually, the amounts of nitrogen and phosphorus
added were greater than required, on the basis of previous experience;
almost one-third more was required to give 50-percent protein. The
additions had been calculated on the expectation of higher BODs to COD
ratios than were actually observed. From this, larger mycelium yields
were expected than were obtained.
Microbial Pattern
Samples of the cultures from the ditch and pool were examined micro-
scopically at regular intervals. Some aspects of culture behavior were
evident on macroscopic observation. Among these were clump size. The
mycelial clumps were never as large as those obtained in laboratory
cultures. They were large enough to be readily visible; some were as
29
-------
10
•o
0>
Q_
"5
•= 0
IN.
O
Q.
10
IS 5
norganic
I
10
20 30
Day of Operation
40 50
Figure 17. Concentration of Phosphates in Feed and Effluent
of Ditch
30
-------
100
80
60
o
E
40
20
10
20 30
Doy of Operation
40
50
Figure 18. Ammonium Ion Concentration of Effluent
from Ditch During Operation on Corn
Canning Wastes
31
-------
large as several millimeters in diameter. The difference from labora-
tory experience in clump size may have been a function of the greater
mechanical violence of the aerator systems or may have been caused by
other factors.
Microscopic examination showed the presence of few yeasts or bacteria
at all times during the tests. Another fungus, believed to be a
Geotviawrn, did begin to appear in the ditch a few days after inoculation,
and by mid-season comprised 50 to 70 percent of the mycelial mass. It
became less prominent in late season, and by the time of discontinuance
of operations comprised not more than 20 percent of the mycelial mass.
An occasional clump of this organism was seen in the pool culture, but
it never became prominent. Some of what was apparently Geotinawn was
present in laboratory fermentations of corn wastes conducted in the
earlier Fungi Imperfecti project.
There was evidence that the mycelial clumps were smaller at the lower
temperatures encountered in the pool and ditch during part of the project,
Recovery of Mycelium
Several techniques were investigated for recovery of effluent mycelium.
The mycelium was too fine for efficient recovery by simple screening; a
process which had been successful in laboratory studies. Neither was
vacuum filtration promising. It, too, had worked satisfactorily on
laboratory cultures. It proved possible on vacuum filtration of labora-
tory cultures to build up mycelial filter cakes 1/8- to 1/4-inch thick
before filtration rates slowed to any marked degree. Using pilot-plant
mycelial material, however, filtration rates were reduced to 0.1 gallon
per minute per square foot of 1/2-inch fungal cake thickness by the time
four gallons per square foot had been filtered.
Considerable investigation was made on the use of a Sweco vibratory
filter for mycelium recovery. Results are shown in Table 3. These
results are considered the poorest that might be expected, since they
were obtained late in the season when the mycelium was especially fine
because of cold weather. The mycelium concentrate from the No. 120
Sweco screen contained 2.5-percent solids. It was filtered readily on a
vacuum filter to produce a cake 0.5- to 1-inch thick and of 20-percent
solids.
The vacuum filter cake formed from fungal material collected on the Sweco
unit could itself be used as a filter to clean up the effluent from the
Sweco. A 0.5-inch fungal cake accommodated 20 gallons of Sweco effluent
per square foot of surface before slowing to a filtration rate of 0.2
gallon per minute per square foot.
32
-------
Table 3. Performance of Sweco Vibratory Filter on
Corn Waste Effluents
Screen
Mesh
94
105
120
165
Gallons Flow
per sq ft
per min
8
6
6
2
Percentage
Mycelium
Recovery
—
—
52
65
COD
Effluent
1352
928
664
332
BOD 5
Effluent
—
—
142
74
Another promising method of cleaning up the filtrate from the Sweco was
by the use of a sand bed. It was found possible to filter 20 bed-volumes
of Sweco 120 filtrate at flow rates of from 2 to 6 gallons per square
foot per minute by using an 8-inch depth of 60 to 80 mesh sand covered
with a 12-inch depth of 16-20 mesh sand and with head depths of less
than 24 inches. The filtrate was reduced from an initial turbidity of
"80" on a Klett scale to a turbidity of "4.5". The latter turbidity
corresponds to a value of 50 by the Jackson candle turbidimeter. The
mycelium contained in the sand bed could be recovered by back washing
with one volume of water.
Settling was investigated as a means of recovering the mycelium, but did
not look promising. A clear supernate was obtained, but the settling
rate was frequently as slow as one inch per hour.
Centrifuging gave rapid clearing at forces above 600 xg.
It was also possible to filter, through Whatman No. 4 paper, with
reasonable flow rates and freedom from clogging by first bubbling air
through the effluent to cause foaming. Finer materials concentrated
in the foam and could be skimmed off. Filtration could then be con-
ducted with much less tendency to clog.
Drying
Drying tests were limited to laboratory trials. It was possible to
achieve drying in a laboratory oven at temperatures up to 110*, to
yield a product of light brown color and little taste. Surface-hardening
was a definite problem with material from the pilot plant and drying
layers of greater than 1/8-inch thickness would seem impractical unless
some precautions, such as humidity control, were taken.
33
-------
Other Corn Waste Studies
First Season Pilot-Plant Operation
During the 1969 season, the aerated pool was operated for eight days
during the corn canning season and then for an additional 22 days using
accumulated corn canning wastes from the lagoon. The pool temperature
did not rise above 17°C during the entire operating cycle and was below
10°C for the last seven days of operation. The data obtained are
summarized in Figure 19. The COD reduction was considerably less than
obtained in the 1970 season; it was in the neighborhood of 70 percent.
The culture maintained itself well with few yeasts, bacteria, and
protozoa being present.
Study of Lagooned Wastes
The inadequate COD removal observed in the first season of operation
may have been caused by low operating temperatures or the use of
lagooned waste. There was no question from gross observations that
changes were occurring in the lagoon so that material drawn from the
lagoon was no longer equivalent to fresh corn wastes. To compare
digestion of fresh corn waste and corn waste drawn from the lagoon,
parallel continuous fermentations were set up in the laboratory and
operated for 30 days. The corn and lagoon wastewater feeds were
transported to the laboratory in five-gallon polyethylene containers and
frozen until used in this experiment. Each culture system was inoculated
with mycelium taken from the pool (1969 season). Each was maintained at
about 24°C, at pH 3.4, and fed to give a retention time of 30 hours. The
COD of the fresh corn was 4320 mg per liter; the effluent COD was 160 to
200 mg per liter, and the mycelium yield about 1500 mg per liter. The
COD of the lagoon waste was 2288 mg per liter; the effluent COD was
about 400 for the first 15 days, and then dropped to about 150, and
remained there. The mycelium yield was 1000 mg per liter. It was
concluded that there were no significant differences in performance
between fresh and lagooned corn wastes that were discernible in this
experiment. These results are shown in Figures 20 and 21. The arrows
in these figures refer to breakdowns in the feeding equipment which
produced both fungal starvation and washout before repairs were made.
In the 1970 season, the question arose as to whether the finer mycelium
and the appearance of another fungal species in the ditch culture might
be a function of use of lagooned feed, rather than feed directly from
the plant operations. To help answer this question, laboratory shake
flask experiments were conducted using fresh plant waste and lagooned
wastes as media, and using either inoculum from the ditch or from
laboratory T. viride culture. Other variations were the use of different
ratios of ammonium sulfate and sodium phosphate. Only trivial growth
34
-------
3000
25r 2500-
20
600
£
3
£
.«
10-
*Feed COD is from the corn canning receiving
lagoon and dropped off linearly during the
30 days of this study.
Figure 19. Performance of Aerated Pool in Continuous
Treatment of Corn Waste by T. viride
35
-------
3000r
o-o—o-o
2500
2000
o>
E
Q
O
O
'500
1000
Theor. COD
Feed COD = 4320 mg/1
? a
o'
500
1 II
—o
I
Fungus —>• \
\
COD
10
15
Days
20
25
1250
1000 -
I"
E
750 >>
o
o>
I
500
250
30V
Figure 20. Continuous Laboratory Digestion by T. viride
of Corn Wastes Drawn Directly from Plant
Effluent
36
-------
U)
o
O
O
3UUU
2500
20OO
1500
1000
500
-0
X* \
— O \ ~~
s \
0
/ \
r p \
/Theor. COD / \
Feed COD - / \ °_ _ O _Q
2288 mg/1 p | ^O °
Feed Rate = / 4 ^
10 ml/min/ [ 2
/
O
A ; ]
\i
I WT»- - 0v^
1 / •-•
I / ^^
0T----° , i.i i i
*j\j\j
250
1000
750
5OO
250
i
n
05 10 15 20 25 30
Days
E
o
o»
c
Figure 21. Continuous Laboratory Digestion by T. vivide of Corn Wastes
Drawn from the Receiving Lagoon
-------
was obtained in any instance using waste from the lagoon. Ten times as
much growth was obtained using fresh corn canning wastes, regardless
of the source of inoculum. The lack of growth on lagooned wastes using
inoculum from the ditch was inconsistent with the fact that successful
removal of BOD5 and mycelium production was regularly being achieved
in the ditch itself on these same lagooned wastes. Subsequent to this
experiment, the point of withdrawal of feed from the lagoon was moved
as close as possible to the point of entry of the plant discharge•
This caused a slight increase in mycelium floe size, but nothing
impressive.
Temperature Effects
Laboratory experiments were undertaken to determine whether low
temperatures were the probable cause of poor COD removal during the
1969 season. The approach was to set up two parallel fermentations,
one operated at room temperature (21 to 24°C) and the other in the cold
room (12°C). Feed rates were varied and mycelium was recycled in the
cold room fermentation in an attempt to increase effectiveness of COD
removal. Results are indicated in Figure 22. The feed to both the room
temperature and the cold room fermentations had a COD of 1735 mg per
liter. The initial feed rate was 10 milliliters per minute, which is
equivalent to a retention time of 30 hours. At the arrow marked with a
"1" on the graph, the feed rate of the cold room fermentation was
reduced to 7.5 milliliters per minute. At the point marked by arrow 2,
recycling of the mycelium was instituted. The recycling returned nearly
all of the mycelium to the fermentation. It is seen that from that
point on, the COD gradually dropped to a level of 200 mg per liter.
Probably some lysis of mycelium occurred since the final level of
mycelium in the fermentation was only moderately higher than expected
with this level of COD removal, even without recycling of mycelium.
The COD level eventually matched that attained at room temperature.
Recycling of mycelium to keep the standing concentration high was
necessary to obtain reasonable COD removal at 12°C.
38
-------
1800
1600-
• i • r • i
o
Figure 22. Performance of Laboratory Treatment of
Lagooned Wastes at Two Temperatures
39
-------
Pea Wastes
Operating Conditions
The pattern of operating conditions used in treating pea wastes is shown
in Figures 23 through 31. In interpreting these figures, the period
from the 1st through the 35th day includes the pertinent experience with
pea wastes; after the 35th day, silage juice was added in amounts that
contributed the greater part of the BODs load.
The retention times used were initially 45 hours, and were stepped down
to 18 hours by the end of the period of treatment of pea wastes. The
pattern of regulation of retention times is shown in Figure 23. Addition
100
M
o
1 50
-------
_ 4
o
O»
O
8
I 3
$
CVJ
o
I
Pea Waste
•"I
Pea Waste I
+Silage
10
Figure 24.
20 30
Days after Inoculation
40
50
(NHlt)2SOit and NaH2POit Additions to Pool in
Pounds per 1000 Gallons of Feed
of ammonium sulfate and sodium dihydrogen phosphate is shown in Figure
24. Addition of phosphate was not begun until the 16th day. The amount
of sulfuric acid required to maintain a pH of about 3.5 is shown in
Figure 25. The amount required was greater than for corn wastes; it
averaged 6.5 pounds per 1000 gallons of feed. The temperature pattern
is shown in Figure 26. The temperature dipped to 16°C on two days, but
was generally above 18°C, and on one day, reached 30°C.
The total load of BOD5 and COD fed to the pool per day is shown in
Figure 27. The general pattern was one of increasing loads as the season
progressed, with the initial BOD5 loads being in the neighborhood of 30
pounds per day and loads late*r in the season being about 50 pounds per
day. COD loading increased from initial levels of about 80 pounds per
day to levels, later in the season, of about 120 pounds per day.
41
-------
30
o
o>
O
O
2 20
20
10
I ' |
Pea Waste I
+ Silage |
I
I
10
20 3O
Days after Inoculation
40
50
Figure 26. Temperature of Pool Contents at 11 a.m. Each Day
42
-------
200
'50
0100
w
o
in
o
o
m
50
I 'I
Pea Waste
10
20 30
Days after Inoculation
40
50
Figure 27. Total Loading of Pool in Pounds BOD5 or COD per Day
Removal of Organic Matter
COD levels in the feed and effluent are shown in Figure 28, and cor-
responding BOD5 levels in Figure 29. COD levels have been reduced from
values in the neighborhood of 1600 to values in the neighborhood of 300.
In eight days, the values were as low as 150 mg per liter. Most of the
peaks and valleys in the effluent COD curve are unexplained, but the
increase on Day 15 corresponds to a change in the floral balance with
the appearance of a much greater proportion of a thin-stranded fungus.
The effluent BOD5 pattern shows a discontinuity at Day 22. Before
Day 22, it vacillated about 20 mg per liter and was consistently below
43
-------
45. BODs removal was above 95 percent during the latter period of
operation.
A more detailed analysis of feed and waste characteristics obtained
from frozen composite samples, which included equal volumes from 20
samples over the course of the operating season, is shown in Table 4.
The ash content of the wastes was very high. Most of this ash was
sodium chloride used in the plant to separate the peas on a density
basis.
2500
2000-
1500
Q
o
o
1000
500
I
Pea Waste
+Silag I
10
20 30
Days after Inoculation
40
50
Figure 28. COD of Feed and Effluent Streams
44
-------
2500
TT
2000
1500
o
ffl
1000
500
Pea Waste
Pea Waste
+ Silage
Pea '
Waste I
10
20 30
Days after Inoculation
40
50
Figure 29. BOD5 of Feed and Effluent Streams
45
-------
Table 4. Analysis of Feed and Effluent Pea Wastes
Test
COD
BOD 5
TOC
Suspended solids
Total solids
Volatile solids
Mycelium (dry wt)
Ash
Ammonia as N
Phosphate as P
Milligrams per Liter
Feed
1650
772
962
264
6818
917
0
5898
45
13
Clarified
Effluent
324
61
126
1*
6585
121
420
5874
5
10
*"Suspended solids" in this instance refers to
suspended materials not removed by filtration,
but which could be collected by centrifugation.
The ratio of BOD5 to COD in the feed averaged about 0.5. The ratio in
the effluent averaged about 0.2, and during the latter part of the
season, was in the neighborhood of 0.1.
Mycelium Recovery
The yield of mycelium is shown in Figure 30. The yield is not a con-
sistent fraction of the BOD5 removed. Between one and 15 days after
start-up, the recovered mycelium had a dry weight equivalent to about
one-third the BOD5 removed. Between 15 and 20 days it increased to
about 50 percent, and by the end of the pea season it was about 60
percent of the BODs removed.
46
-------
^ 1.0
o»
0.5
a>
u
0
I 'I
Pea Waste
i ' •
I Pea Waste
+ Silage
I
I
TT
10
20 30
Days after Inoculation
40
50
Figure 30. Dry Weight in Grams per Liter of Solids
Filtered from Effluent Stream
Oxygen Consumption
Dissolved oxygen levels are represented in Figure 31. According to the
manufacturer, the aeration unit used provided aeration at the level
expected of a 1.5-hp floating aerator. It kept up with the oxygen
requirements of the system at all times during the treatment of pea
canning wastes. The lowest DO values were 2.0, and generally the
values were above 3.0 milligrams per liter in both pea and corn fermenta-
tions .
Microbial Pattern
The microbial mass in the pea waste treatment was always dominated by
fungi, with bacteria and protozoa being present only in small numbers,
On about the eleventh day there was a shift in fungal type from
Tridhoderma viride, which had been used as an inoculum, to another
fungus, possibly of the Fusariw genus. T. vivide continued to be
evident but in decreasing amounts until, at the end of the season, it
formed probably not more than 10 percent of the mycelial mass. The
FwaoHum-like organism produced much finer floe than did T, viride
and so could not be recovered on coarse filters. It could, however,
be recovered on filter paper (Whatman No. 4).
47
-------
12
8
i
>s
X
O
TJ
Q) .
.> 4
O
Pea Waste
I
I r
Pea Waste
+Silage
10
20 30
Days after Inoculation
40
50
Figure 31. Dissolved Oxygen Concentration in Pool at 11 a.m. Each Day
A laboratory study was undertaken to determine whether the changes
that occurred in the lagoon from which the feed to the pilot plant was
drawn were of prime importance in affecting the microbial balance in
the aerated pool. It was found in shake-flask studies that T. viride
grew well on fresh plant waste and poorly on wastes drawn from the
receiving lagoon. Using a mixture of the Fusorii^n-like fungus and
T. viride as inoculum, the Fwscwiitffl-like organism dominated when
wastes from the receiving lagoon were used as substrate. Additions
of ammonium sulfate; ammonium sulfate, sodium dihydrogen phosphate,
magnesium sulfate, calcium chloride, ferrous sulfate, manganous sulfate,
zinc sulfate, and cobalt chloride; starch; dextrose; or peptone did not
change the dominance of the Fueoi'iurn-like organism when the wastes from
the receiving lagoon were used. Also, the closer to the point of plant
discharge that lagoon samples were taken, the better they supported
T. viride growth.
48
-------
Corn Silage Wastes
In the corn processing operation at Glencoe, the husks and cobs are
accumulated in a large pile which undergoes anaerobic fermentation.
An aqueous solution called "silage juice" is released and stored in a
lagoon. The solution has an extremely high BOD5 level and presents a
difficult disposal problem. Near the end of the pea canning season,
the BOD5 of the pea canning waste from the lagoon dropped to a low
level and presented an opportunity to test the effectiveness of the
Fungi Imperfecti, system on the silage wastes. The silage juice from
the previous season was run into the pilot plant with the pea waste
at this time. The operating variables and performances are represented
as extensions of the corresponding data on pea canning wastes in
Figures 23 through 31. Figures 28 and 29 show the proportion of the
total load from pea canning wastes and the proportion from silage
during the period when silage Juice was being added to the fermentation.
The amount of COD or BOD5 from silage juice is the difference between
the dotted lines which show the levels from the pea waste and the solid
lines which represent total levels fed to the fermentation.
The levels of COD in the silage juice were 40,300 mg per liter and
the BOD5 level was 38,000 mg per liter. In calculating the COD and
BOD5 levels being added to the pool, account was taken of the volume
of pea wastes that were being added coincidentally and which acted as
diluents.
The retention time during treatment of silage juice was 18 hours
(Figure 23). The temperature was above 20°C on all days (Figure 26).
Sulfuric acid requirements were slightly higher than on pea wastes
alone. They averaged 8.1 pounds per 1000 gallons of wastes treated
(Figure 25). Ammonium sulfate and sodium phosphate additions were
made (Figure 25) assuming the levels in the silage juice to be low.
Later analysis showed nitrogen levels to be high enough (4.04 percent
of the total solids) to support a mycelium of 50-percent protein,
assuming 50 percent of the solids are recovered as mycelium.
COD removal (Figure 28) was about 83 percent; BODs removal (Figure
29) was about 95 percent during the first five days of silage addition.
This later decreased to 80 percent. The decrease in performance was
associated with a decrease in DO levels to near zero. Calculating
oxygen use at the times of low DO, and assuming that the 1.5-hp unit
provides three pounds of dissolved oxygen per hour, one arrives at an
estimated oxygen use of 72 pounds per day. About 136 pounds of BOD5
were being removed per day. The calculated requirement is therefore
0.5 pound of dissolved oxygen per pound of BOD5 removed. The mycelium
yield appeared to be low, equivalent to .only 25 to 30 percent of the
BODs removed. The biomass continued to be dominated by the Fusariwn-
like organism that dominated the pea waste fermentation.
49
-------
Mechanical Performance
There was some concern in designing the systems that there might be
dead spots where the flow rate was too low to keep the mycelium in
suspension and where oxygen levels would be much lower than the average
for the system. Sampling of the pool at a time when DO levels were
below one mg per liter gave the pattern shown in Figure 32. The level
of total solids represented by the COD on the unfiltered samples, showed
uniform suspension at the sampling points. The DO values differed, but
only over a twofold range. Flow velocities differed by 2.3 fold between
extremes among the points examined. These examinations were made with a
1-hp aerator. With a 2-hp aerator, the circulation was obviously
adequate. The values recorded in Figure 32 are averages of three samples
taken at each sampling site.
The circulation pattern in the ditch was even better than in the pool.
With the oversized aerator used, the flow velocity was three feet per
second. There was a tendency for the mycelium to concentrate in the
foam on both units. Advantage could be taken of this tendency in
harvesting mycelium.
50
-------
Figure 32. Flow Velocities, Dissolved Oxygen and COD Values at Selected
Sampling Points in the Aerated Pool. (Sampling points are
indicated as ® .)
-------
SECTION VI
DISCUSSION
General Effectiveness
Organic Compound Removal
The general effectiveness of the Fungi Impevfeoti digestions on the
three wastes to which it was applied in the pilot-plant studies is
summarized in Table 5. The figures used in preparing the Table are
neither the best nor the worst that could have been chosen from
performance data, nor are they general averages; rather they represent
averages from periods of stabilized favorable performance. BODs removal
was good and COD removal fair, in the cases of both corn canning wastes
and pea canning wastes. In neither instance was removal as good as that
obtained in laboratory fermentations where BOD5 removal ran above 99
percent and COD removal about 96. The reason for the poorer performance
is probably in the finer mycelium produced in the pilot-plant operations.
Samples for analysis were prepared by filtration of a relatively small
volume of effluent through Whatman No. 4 filter paper. Because the
mycelium was fine, this procedure did not produce an entirely clear
effluent. In the laboratory, a much larger mycelium was produced which
was more easily retained on the filter. An indication of the validity
of this reasoning is seen in data cited in Table 6 in which a composite
effluent was completely clarified by centrifugation. The COD was
reduced from 305 to 215 and the BODs from 59 to 41. These levels in the
effluents would correspond to 92-percent reduction of COD and 97.5-percent
Table 5. General Efficiency of Fungi Imperfeoti Process
Percent BODs removal
Percent COD removal
Percent TOC removal
Mycelium produced per unit
BOD 5 removed
H2SOit use — lb/1000 gallons
Retention time — hours
Corn Canning
Wastes
96
88
93
0.5
4.0
22
Pea Canning
Wastes
95
81
87
0.6
6.5
18.
Silage
Wastes
80
83
85
0.3
8.1
18
53
-------
Table 6. Characteristics of Composite Samples of
Corn Waste Feed and Effluent Collected
in Mid-Season**
COD
BOD 5
Suspended solids
Total solids
Volatile solids
Mycelium (dry wt)
Ash
TOG
Ammonia, as N
P04E
Milligrams per Liter
Feed
2536
1580
210
2520
1490
0
1070
1608
49
9
Effluent
305
59
101*
2237
521
646
1058
104
23
0.1
Clarified
Effluent
215
41
0
2136
432
0
998
81
0.2
0.05
*"Suspended solids" in this instance refers to suspended
materials not removed by filtration, but which could be
collected by centrifugation.
**Composites are from the same 20 days in each instance.
reduction of BOD5. It is believed that similar improvements in removal
of BOD5 and COD could be effected in the case of pea canning wastes, but
it is not certain that they could be achieved in the case of silage
waste. The conversions of BOD5 to mycelium in the case of corn and pea
canning wastes are not greatly different from those obtained in
laboratory digestions.1 They are slightly lower in the pilot plant, but
this may represent loss of fine mycelium. It would be of interest to
determine if the mycelial yield could be improved by shortening the
fermentation time, to reduce losses by endogenous respiration, and by
decreasing the amount of nitrogen added to the treatment system. Higher
levels of nitrogen generally promote more oxidation of carbon sources.5
On the other hand, reducing added nitrogen to the point of reducing the
protein content of the mycelium would defeat one of the objectives of
54
-------
the use of Fungi Imperfeoti. This objective is that of using the fungus
as a high protein animal feed. Analysis of a sample of fungus obtained
from the corn waste digestion showed a nitrogen content of 8.63 percent.
This corresponds to 54-percent protein if one multiplies by the con-
ventional conversion factor of 6,25, There is, of course, no assurance
that all of the material recovered by filtration is mycelium; some
could represent undigested suspended solids present in the original
feed. The level of suspended solids in the feed was, however, low.
It averaged only 210 milligrams per liter in the corn canning waste and
50 milligrams per liter in the pea wastes.
Nitrogen and Phosphate Removal
Nitrogen and phosphate removal is of particular interest because it is
highly desirable, from the standpoint of water quality, that the levels
in the effluent be very low. In the 1970 season of operation, more
ammonium sulfate and sodium phosphate were inadvertently added than
required either by calculation or by previous experience. Even so, the
levels of phosphate were uniformly lower in the effluent than in the
influent, and at several periods were essentially zero. Nitrogen
levels were likewise essentially zero at some periods of operation.
These low levels had no apparent effect on BOD 5 or COD removal. These
observations, coupled with observations in the laboratory, make it appear
highly likely that the process can be operated with almost no leakage
of inorganic nitrogen or phosphorus into the effluent stream. The
levels of organic nitrogen and phosphorus will be largely a function
of the amount of mycelium that escapes into the effluent stream.
Acid Usage
Acid usage was disappointingly high, and on the basis of requirement per
pound of BODs processed, was almost three times as high as in laboratory
experience with corn processing waste. This may only reflect a higher
level of hardness in the water and a lower BODs concentration; i.e.,
more water has to be acidified per unit of BOD5 handled. Another
important factor bearing on acid use is the effect of anaerobic changes
that take place in the lagoon. Direct titration of fresh wastes and of
lagoon wastes showed that almost twice as much acid was required to
titrate the material drawn from the lagoon as was required to titrate
the fresh plant wastes. The acid actually required in the treatment was
about two-thirds that required to titrate lagoon wastes; thus giving
evidence of some acid production by the digestion itself. It is quite
evident that a variety of changes occur quickly in the lagoon. These
are evidenced both by the effect of the lagoon materials on Triohodezma
viride growth and by changes in smell and color.
Closely related to the question of acid use is the question of the
amount of base that may be required to return the effluent stream to
neutrality. Titration of the effluent streams from treatment of corn
processing wastes and pea processing wastes showed return to neutrality
55
-------
could be achieved by use of 0.0041 equivalents of base per 1000 gallons.
Comparing on the equivalency basis, only about five percent as much base
is required to return the solution to neutrality as was required of acid
to effect the acidification.
Retention Times
Retention times were approached conservatively in attempts to achieve
good BOD 5 and COD removal. After long enough operating times at one
retention time for the treatment to become stabilized, further decreases
in retention time were made. In no instance did decreases in retention
time increase BODs levels in the effluent. It, therefore, seems likely
that retention times were never a limiting factor in performance, and
that much shorter times might have been used successfully than were
tried. It would be of great interest to determine the effect of re-
tention time on growth habit of the fungi and upon mycelial yield per
unit of BODs removed.
Oxygen Requirements
Knowledge of oxygen requirements is critical to estimates of the economy
of the process. Estimates achieved during the present study leave more
uncertainty than is desirable. Estimates of oxygen requirements made
during the operation on corn wastes were 1.2, less than 0.8, and 0.7
pounds of oxygen per pound of BODs removed. These values differ
markedly from the value of 0.14 pound of oxygen per pound of BODs
removed estimated during the laboratory phase of Fungi Imperfecti
studies.1 Some of the measurements were by similar techniques and no
errors in calculation could be found. Perhaps it was not valid to
assume that readings of oxygen consumption at a selected point in time
were representative of the average oxygen consumption rate over a 24-
hour period. In any case, it seems certain that the estimate of 0.14
pound of oxygen per pound of BODs removed is an impossibly low figure.
Removal of one pound of BODs yielded not more than 0.6 pound of mycelium.
The missing 0.4 pound must have been completely oxidized. It is reason-
able to assume that the BOD5 was not originally at a higher stage of
oxidation than carbohydrates and that it was oxidized to carbon dioxide
and water. This would require 0.4 pound of oxygen on stoichiometric
considerations alone. Allowing only for this portion of oxygen con-
sumption would require the use of 0.4 pound of oxygen per pound of BOD5
removed. Extrapolating from this base, a figure of 0.7 pound of dis-
solved oxygen used per pound of BOD5 removed would seem the most
favorable estimate that one can presently make. No estimates of oxygen
use on pea canning wastes can be made. Estimates on silage wastes were
0.5 pound of DO used per pound of BOD5 removed. It would be of consider-
able interest to alter such variables as retention time and nitrogen
addition in the hope that they would increase mycelium yield. This
should decrease oxygen requirements, if more carbon is transferred to
cell mass.
56
-------
Temperature Effects
Low temperatures definitely limit the effectiveness of the process.
During the first season, temperatures below 12°C had considerable
adverse effect. During the 1970 season, temperatures down to 10°C were
successfully accommodated. Successful operation at still lower tempera-
tures could probably be achieved if part of the effluent mycelium were
recycled. It is also possible that on prolonged operation at low tempera-
tures, selection would take place for substrains that grow rapidly at low
temperatures. With depression of growth rate by low temperatures, a
finer mycelium was produced. In some instances, it may be possible to
use waste heat from processing operations to maintain the temperature of
the treatment facility in the winter months.
Microbial Stability
The microbial system remained fungal in type at all times and it is
thought that acidification was adequate to give dominance to a fungal
type flora. This generalization was derived from common experience with
the effect of acidity on flora in sewage treatment'' as well as from our
own experience with Fungi Impepfeoti. There were, however, shifts in
the type of flora that were dominant. Laboratory experience led us to
believe that a FusaPium type of fungus would predominate on pea wastes,
and such indeed proved to be the case even though inoculation was with
Triohoderma viride. Trichoderma viride remained dominant during the
short 1969 season operation on corn waste. It also remained dominant
in the aerated pool during the 1970 season, but became secondary to
Geotricum in the ditch during midseason operations. One reason for the
submergence of T. v-Liride to Geotricum may well have been the effect of
the lagoon waste. In laboratory shake-flask studies, lagooned wastes
proved to be unfavorable to Triohoderma viride growth. One argument
against this theory is that moving the point of intake from the lagoon
nearer to the point of plant effluent discharge had no immediate effect
on the microbial balances in the ditch. The percentage of Trichoderma
viride did however gradually increase until it was once again predominant
late in the season. Another factor of importance in determing which
fungus predominated in the 1970-season operation may have been excesses
of nitrogen and phosphate. A tentative generalization can be made that
acidification favors fungal growth, but that the fungus which will
predominate will depend on the composition of the waste.
Harvesting
The finer mycelium encountered in the pilot plant, compared with the
mycelium seen in the laboratory, made harvesting a more critical problem
than anticipated. The best combination explored seemed to be that of
using a gross technique such as the Sweco filter to remove as much of the
mycelium as possible, and then use of a sand filter to remove the finer
57
-------
mycelium that passed through the filter. Material recovered by back-
washing the sand filter could probably be further concentrated by
recycling it through the Sweco filter. The concentrate from the Sweco
filter could readily be concentrated to a 20-percent solids content by
vacuum filtration. Ideally, one should vary operating conditions to
attain the coarse, flocculant mycelium observed in the laboratory
fermentations. Work should be done to determine the role that violent
agitation plays in producing a finer mycelium. There was also evidence
that higher temperatures favored a coarser mycelium. Retention time
may play a role and this should also be investigated. There was
evidence, at least in shake-flask experience, that the use of lagooned
waste promoted finer mycelium as well as promoting selection of species
that inherently produced finer mycelium.
Inoculation
The inoculation techniques used generally worked well and Triohoderma
viride initially established itself in all cases even though it was, on
occasion, displaced by other Fungi Imperfecti. The only failures
occurred when pH control failed immediately after inoculation. The time
required for the fungus to begin vigorous removal of BOD5 after inocula-
tion was generally longer than desirable. Four to six days were required.
This parallels several laboratory experiences. It may be that some
substrain selection is required in the continuous culture before it
really begins performing vigorously. If this is true, then substrain
selection made during inoculum buildup should shorten the initial lag
in waste digestion start-up.
The expedient of producing inoculum in a culture rich in nutrient to
obtain a high concentration of mycelium has been successful. Preserving
mycelium in the cold, in a frozen stage, or after dehydration allowed
regeneration of the culture, but only after some difficulty; the
yeast contained in the mycelium in relatively low numbers multiplied
much more quickly than the fungus and dominated the cultures for the
first few days.
Mechanical Performance
In planning the experimental program, it was hoped that comparisons
could be made between the floating aerator and the ditch-cage aerator
system for use with Fungi Impevfeoti. For several reasons which included
discrepancy in size of the aerator units, differences in floral balance
in the two systems, and a loss of effective pH control in the pool at a
critical time in the season, comparisons x>f the two systems was difficult.
Both worked satisfactorily. Slightly better performance was obtained
in the ditch, but it is not certain that this was-caused by the mechanical
factors associated with this configuration. The slightly higher tempera-
ture that prevailed in the ditch is considered a slight advantage.
58
-------
Because of the use of acid conditions, evidences of corrosion were
watched for. The cage aerator was coated with an acid-resistant paint
and no evidences of corrosion were observed. The shaft of the floating
aerator unit did show corrosion. Two unusual factors probably con-
tributed to this. One was because of a loose connection in the pH
control equipment, the pH of the pool unit dropped to well below 2 for
several days. The other was that 316 stainless steel was used for the
bulk of the unit and 304 stainless steel for the propeller shaft. This
would tend to promote corrosion of the shaft.
Other equipment generally performed satisfactorily. A loose connection
in one of the pH controller units caused disasters on two occasions, but
this cannot be considered characteristic of the equipment. It was
necessary to clean the pH electrodes frequently to obtain reliable
performance. Otherwise they became covered with fungal growth which
slowed their responses.
Comparisons with an Extended Aeration Process
Another experimental program was run independently by the Green Giant
Company in parallel with the Fungi Imperfeeti studies. It is described
below.
An aerated lagoon for treatment of cannery wastewater was con-
structed at the Green Giant Company, Glencoe, Minnesota plant during
the summer of 1969. In relation to the total wastewater load, the lagoon
is of pilot scale and was built to provide an actual field evaluation with
extended operation, as applied to vegetable canning factory waste. The
experience reported here covers only a partial season.
Description of the Facilities
The aerated lagoon consists of a triangular-shaped earthen pond having
a liquid capacity of 1.35 million gallons. The water surface is
approximately 200 feet along each leg of the triangle. The liquid depth
is 15 feet. The earthen sides slope up from the bottom a distance of
13 feet vertically and 26 feet horizontally (2:1 slope). The sides
continue to slope upward from that point a distance of 2.5 feet
vertically and 15 feet horizontally (6:1 slope). From the 15.5-foot
level to the top of the berm at 18 feet, the slope returns to the 2:1
ratio. Thus,the earth slopes on the inside of the ponds are at 2:1
with a 15-foot wide berm at the water line which has a flatter, 6:1
slope.
The lagoon is aerated by one 75-hp platform mounted "Lightnin" aerator,
Model LAR-150, manufactured by Mixing Equipment Company, Rochester, N.Y.
The unit consists of a 75-hp horizontal-shaft electric motor connected
by a flexible coupling to a right-angle speed reducer. The speed
59
-------
reducer unit has a 5-inch vertical steel shaft extending from the plat-
form to the water surface. At the end of the shaft is an impeller
consisting of four blades of flat one-inch steel plate. The diameter
of the impeller is 9ft, 6 inches. The shaft and impeller rotate at
37.9 rpm. The impeller is positioned at the water surface with the top
of the blades just barely covered when in a static position. Except
for the four steel blades and the cast steel hub, there is nothing
submerged in the water. The aerator is suspended over the lagoon by
a structural steel platform supported by three reinforced concrete
columns with concrete footings. The lagoon bottom directly beneath the
impeller is covered with a 6-inch thick concrete slab extending around
the column footings, to protect against the possibility of erosion
caused by the action of the impeller. There is a structural steel
footbridge from the lagoon bank out to the support platform.
Wastewater is fed into the lagoon through an orifice tank. This consists
of a steel tank with a bolt-on plate with circular orifice. There is an
overflow weir at the top of the tank with any excess feedwater sent
back to the source; thus providing an orifice under constant head. The
rate of discharge into the lagoon is controlled by providing an orifice
of given diameter. The formula for orifice discharge used is:
Q = 11.94 d2t/H
where Q = discharge in gpm
d = orifice diameter in inches
H = head on orifice in feet
The formula assumes a coefficient of discharge of 0.6. There has been
considerable investigation of the hydraulics of orifice discharge at the
size and head range involved, and this formula is considered accurate to
within one percent or less. It is, thus, the most accurately measured
variable involved in the project. The orifice tank is provided in
duplicate to control the simultaneous discharge of two separate waste-
waters into the lagoon, if such were a desired part of the experi-
ments. In addition to an influent feed pump, a second pump is provided
for recirculation of treated water back into the lagoon.
The discharge of wastewater out of the lagoon is controlled by a 51-inch-
diameter steel weir plate overflowing into a catch basin. The catch
basin discharges through a culvert to the adjacent lagoon No. IB.
Figure 33 provides a schematic view of the facilities.
60
-------
N
Scole-f=200'
Influent Feed Pump
Approx 3OO GPH'
Capacity
16" Force Main
From Fxtory
Orifice Bo
75 HP Surface Aerator
Pond 'IB
Holding Pond for Pilot Lagoon
4.5ft water depth-18.15 MG
(Future Aeration Lagoon)
Silage Liquor Retention
9.5 M.G.
-300 GPH Recirulation /
Pump /
/
! /
Figure 33. Pilot Lagoon for Extended
Aeration at Glencoe, Minnesota
J
61
-------
Evaluation of Performance
Even though the aerator did not begin operating until September 6, it
was possible to obtain quite a bit of information. The waste feed
supply was taken from the original Pond No. 1 of the seven staged ponds,
and thus the system continued to operate even after the end of pea and
corn pack. There was no observable difference in the biological process
whether the waste was treated almost immediately fresh from the factory,
or if it had been retained in Pond No. 1 for a long period of time.
The system is evaluated as to the following features:
Mechanical Reliability of Aeration Equipment. The aerator functioned
continuously from September 6 to November 14, with no problem or
difficulty. The machine operates unattended although someone visits
the site once a day. No maintenance or attention was required except
for change of gear lube at the end of the break-in period.
Cold Weather Operation. The equipment can be operated in cold weather
provided incoming waste furnishes a heat source, or until the ambient
temperature and the mixing aerating action turns the entire pond to ice.
However, splashing of water onto the structure must be controlled to
prevent ice formation on the platform. This pilot installation was
found to be deficient because water is splashed against the face of the
supporting columns at such an angle that some water is thrown up onto
the platform. It is believed that this can be partially corrected by
adding a baffle or deflector at each column. The unit was shut down
November 15 when the temperature dropped to 12°F.
Lining of Earth Lagoon. Because the BOD5 concentration of corn waste-
water is quite high (about ten times as high as domestic sewage), the
oxygen requirements are also high. This requires a high energy input
into the lagoon. The pilot installation has 55.6 hp per million
gallons of lagoon capacity. This hp-volume ratio approaches the level
required in activated sludge plants where aeration is performed in
concrete tanks. It was the opinion of most of the persons consulted
who had experience with aerated ponds that a concrete or bituminous
lining would be required in the earth lagoon to control erosion of the
banks. The lagoon was constructed without a lining, relying on the
6:1 "broken back" berm section at the water line to control erosion.
This has been completely successful to date with only very slight erosion
at the waterline. This is a significant finding since a lining would
involve great expense.
Biological Efficiency of the Process. The biological efficiency of the
process (i.e.t whether corn waste would actually respond to treatment in
this type of system), was evaluated in the following ways:
A. The efficiency of the process is calculated by the formula:
v - 230 kt
1 + 2.3 kt
62
-------
where: E = percent BODs removed
t = lagoon detention time in days
k = reaction coefficient
During operation, the actual efficiency of the process is determined by
sampling and BOD5 analysis of the influent and effluent from the system.
From the data obtained it has been calculated that the reaction co-
efficient at 20°C is k2o - 0.86. Thus, at 20°C and a 6-day detention
time, BODs reductions of 92 percent are obtainable. This corresponded
in practice to treatment of 44 pounds of BOD 5 per hp per day. It should
be pointed out that the BOD5 tests performed were on the settled effluent
from the system. In practice it is expected that the settled solids on
the bottom of the first pond following aeration will decompose and cause
some subsequent rise in the BOD5 levels. This is not reflected by the
test data.
The limits of performance of the Funai Imerfect'i system, although not
adequately probed, are given for comparison. During one period of
several days (Days 22 to 35 » Figure 24) in the pool, treatment of 50
pounds of BODs per hp per day with 95-percent removal of BODs was
observed. Temperatures during this period varied from 16 °C to 8°C.
B. In addition to detention time, the efficiency is affected by
temperature. The reaction rate coefficient was found to vary with
temperature according to the van't Hoff - Arrhenius equation:
f-20
9 (where 6 = 1.035)
6, the temperature correction factor, is 1.072 in batch systems. This
does not fit the data from continuous systems. By trial and error we
found that 6 - 1.035 best fit these data.
A continuous record of water temperature and air temperature was
obtained by using a soil-air thermograph. From the data obtained, it
was possible to compare the observed efficiencies with the theoretical
calculated efficiencies as temperature and feed rate varied. Some of
the results are shown in Figures 34, 35, and 36.
C. Another variable which affects the reaction coefficient, kf is
the availability of nutrient for the microorganisms to grow and multiply.
From September 16 to October 14, one pound of nitrogen was added to the
system for each 30 pounds of BODs applied, and one pound of phosphorus
was added for each 150 pounds of BODs applied. For a short period, from
September 23 to October 1, the BODs application was increased without
changing the nutrient addition. On October 14, the nutrient addition was
63
-------
CT>
2200
2000
1800
1600
1400 -
_ 1200
Q
O
CD
1000
800
600
40C
20
of Influent, Lj (mq/l)
, L0 (mq/l)
6 8 10 12 15 17 19 22 25
September 1969
I 2
79 14
October 1969
30
Nov.
Figure 34. BOD5 Sample Results for Pond
-------
5.0
Dissolved Oxygen History
25
20
15
o
£
3
I 10
5
0
L Added IQO'Vl a 20*'P per doy
Added 50* N 8 10* P per dov
Temperature History
100
90 -
SC -
70 -
60
Removal Efficiency
Ec = Theoretical Efficiency Based on k2Q=0.86 and 8=1.035
£o - Efficiency Based on BODg Test Results
10
15 20 25
September 1969
30
10 IS 20
October 1969
25
30
5
Nov.
Figure 35. Dissolved Oxygen, Temperature, and Removal
Efficiency in Pond
65
-------
100
90
80
70
O
CD
o 60
o
E
£T 50
"c
I 40
&
30
100
E for-6 day Det. Time
Effect of
Temperature
on
Process Efficiency
kt = k20(1.035>
t-20
kt = k2o(IO35)
t-2C
t,°c
1
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
kt
0.45
0.46
0.48
0.50
0.52
0.53
0.55
0.57
0.59
O.6I
0.63
0.65
0.67
0.70
0.72
0.75
0.78
0.80
0.83
0.86
0,89
0.92
0.95
0.99
1.02
Percent BODs Removal
vs.
Detention Time
Corn Plant Waste
Detention Time, Days
Figure 36. Effect of Temperature and Detention Time on
Removal Efficiency in Pond
66
-------
cut in half. The effects of these changes in nutrient level are not
evident from the data available. More study will be required to deter-
mine the minimum required amount of nutrient.
Aeration Capability of Mechanical Equipment. The biological reaction
coefficient determines the detention time and thus, the pond volume,
assuming a given level of performance is required. Since the difference
in cost of earthwork in building a lagoon of 20 mg versus one of 15 mg
is not large, some variation in the reaction coefficient does not have
a major effect on capital costs. The major part of the capital costs
result from the installation of mechanical aeration equipment. There-
fore, it is important to know the capability of each aerator to handle
BOD5 loading under the conditions anticipated.
Based on the guaranteed capacity of the aerator to supply 5856 pounds
of oxygen per day to tap water under standard operating conditions (3.2
hp per hr), the calculated capacity under lagoon conditions was estimated
at 3860 pounds per day. Therefore, the project was set up to apply 4000
pounds of BODs per day to the aeration lagoon in the initial runs. As
it turned out, the BODs concentrations in the raw waste were less than
anticipated, and the actual application was about 3000 pounds per day.
This feed rate did not tax the capacity of the aerator to determine its
maximum capability. On September 24, the feed rate to the system was
increased to apply 4800 pounds BOD5 per day. After seven days at this
loading, the dissolved oxygen level in the system was down near zero.
The lagoon and effluent samples from the lagoon were odorous, whereas
no odor had been observed before. The settling characteristics of the
effluent solids were lost. It appeared that the aeration equipment
was overloaded, and the load was reduced to its previous level. After
a couple of days the system recovered.
With a system of six-day detention time, a delay is required to
determine the effect of changes in loading. There is an additional
delay for the BOD5 test. When corn and pea pack ended the frequency
of sampling was cut back to once a week or less because of the loss of
seasonal help in the Food Science Laboratory. With the observations
of other variables providing effective data, there was a reluctance to
change too many variables at once. As a result, we have not yet
determined the exact BOD5 removal capabilities of the 75-hp aerator.
Based on an interpolation of the data thus far, it seems safe to
assume that 3300 pounds of BODs Per day can be removed by the pilot
plant without depleting the oxygen. To accommodate the entire waste
stream from the Glencoe plant, approximately ten 75-hp aerators would
be required.
Electrical Power Consumption. The electric utility bill for the
period of September 15 to October 15, was $667. About 90 percent of
the power was consumed by the 75-hp aerator. Thus, it appears to
cost $20.00 per day for electrical energy to operate the aerator.
If the aerator is kept loaded at an overall efficiency of 90 percent,
the BOD5 removal would average 3000 pounds per day. The electrical
energy cost would be $0.0067 per pound of BODs removed.
67
-------
The total BOD5 load at the Glencoe Plant is estimated as follows:
Pea waste - 60 mg at 1,200 mg/1 = 600,000 Ib BOD5
Corn Waste -120 mg at 1,600 mg/1 = 1,600,000 Ib BOD5
Silage juice - 5 mg at 60,000 mg/1 = 2,500.000 Ib BOD5
Total 4,700,000 Ib BOD5
For 90-percent BOD 5 removal in the aeration system the total annual
electrical energy cost is estimated as follows:
4,7000,000 Ib x 0.9 x $0.0067 = $28,200
For peas, this amounts to about $0.0024 per case. However for corn,
including the silage liquor, this calculates out to about $0.0068
per case.
On the basis of increased cost per case as estimated above, all other
operating costs (maintenance, etc.) are negligible.
68
-------
SECTION VII
ECONOMIC ESTIMATES
Estimated costs of operation of the Fungi- Imperfecti system are higher
than in laboratory studies1 for three reasons. One reason is the higher
sulfuric acid demand; another is revision of estimates of oxygen
requirements; and the third is the greater difficulty in filtering
material produced in the pilot plant. The higher sulfuric acid require-
ments are a function of the use of harder water and of a more dilute
plant-waste stream. Amortization costs and interest costs would be
reduced by fourfold per unit of product or per unit of BODs if the
equipment were used the year around instead of for three months.
Cost estimates per pound of product and per pound of BODs treated are
given in Table 7. In making the estimates, it has been assumed that
Table 7. Cost Estimates on Corn Processing
H2SOn
(HNi^SOi,
NaH2POit
Aeration
Power
Investment
Labor
Filtration
Sweco
Sand bed
Vacuum filter
Drying
Heat
Investment
Total
Credit for dry solids
Net Cost
Cents per
pound product
0.86
0.84
0.20
0.78
2.32
1.67
0.18
0.92
0.66
0.35
0.92
9.70
-3.50
6.20
Cents Per
pound BOD 5
0.43
0.42
0.10
0.39
1.16
0.83
0.09
0.46
0.33
0.18
0.46
4.85
-1.75
3.10
69
-------
the BODs concentration in 1.0 mgd of feed is 1600 milligrams per liter,
In calculating sulfuric acid cost, it has been assumed that sulfuric
acid will cost 1.7 cents per pound. Also, it appears that it will be
necessary to use 3 pounds per 1000 gallons of feed if fresh feed is used,
rather than the 4 pounds that were necessary when waste from the lagoon
was treated. In calculating ammonium sulfate costs, a cost of 2 cents
per pound has been assumed for the ammonium sulfate and a requirement of
2.5 pounds per 1000 gallons of feed has been assumed. This is less than
the 3.5 pounds used during the bulk of the operation in 1970, but is
adequate to give a mycelium with a protein content of 50 percent and is
in line with previous experience on requirements. In calculating the
sodium phosphate costs, it has been assumed that sodium dihydrogen phosphate
will be available at 11 cents per pound and that 0.11 pound per 1000
gallons of feed will be adequate. This again is less than used in the
1970 operation, but was adequate in previous experience and should assure
a. near zero phosphate level in the final effluent. In calculating
aeration costs, it has been assumed that 0.7 pound of dissolved oxygen
will be required for each pound of BODs removed, that 1 hp delivers 2
pounds of dissolved oxygen, and that electricity will be available at
1.5 cents per kilowatt-hour . In calculating investment costs, it has
been assumed that $500 per horsepower will pay for both aeration equip-
ment and auxiliary equipment, including the lagoon; that interest costs
will be 7 percent; that the investment would be amortized over 10
years; and that the unit will be operable for 90 days of the year. In
calculating labor costs, it has been assumed that one man can take care
of the unit and that $100 a day will be adequate to cover this item.
In calculating filtration costs it has been assumed that a combination
of a Sweco unit, a sand bed, and vacuum-filter would be required. For
the Sweco unit it has been assumed that a 40-square-foot unit would be
required. Such a unit would cost about $7000. A sand bed to handle
one million gallons a day has been assumed to cost $35,000. A vacuum
filter to dewater 6000 Ib of solids per day has been assumed to cost
$25,000. A drum drier to handle 6000 Ib a day will probably cost about
$35,000. The power cost for drying from 80 percent to 10 percent moisture
has been estimated at 0.35 cents per Ib of dry product, on the basis of
heat at a total cost of 45 cents per million Btu, as follows:
Drying from 80 percent to 10 percent moisture requires evaporation
of 3.89 Ib water per Ib dry product. It is assumed that the
overall drying efficiency is such that 2000 Btu is needed to
evaporate 1 Ib of water. Therefore, the cost per Ib of dry
product is:
3.89 Ib H_0 x 2000 T|^-» x °'45 - 0.0035
2 Ib H0
These costs are tabulated in 'Table 7. Sale of the product might be
expected to return 3.5 cents per pound by analogy with the selling price
of soy meal, comparable in protein content and quality.
70
-------
Among the factors that could lower operating costs per pound of BOD5
in a significant degree are higher BOD5 concentrations, longer operating
seasons, softer water, wastes containing adequate nitrogen and phosphorus,
and a mycelium that filters as readily as did the laboratory materials.
Operating on a year-round basis would reduce investment costs to one-
fourth those listed, reducing the total cost to about 3.0 cents per
pound of BOD5. Production of a readily filterable material, as was
accomplished in the laboratory, would eliminate the need for the sand
filter, to give a cost (on a year-round basis) of about 2.9 cents per Ib
of BODs. The credit of 3.5 cents per Ib of product, or 1.75 cents on
a pound of BOD5 basis, would reduce the net cost to about 1.1 cents per
Ib of BOD5 treated. Further economies would be possible if oxygen re-
quirements, or costs, and chemical use could be lowered. A major re-
duction would occur if drying could be avoided; e.g., by direct feeding
of the wet mycelium. On a year-round basis, a savings of about 0.30
cents would result, reducing the cost per Ib of BOD5 to about 0.8 cent.
71
-------
SECTION VIII
ACKNOWLEDGMENTS
Special credit is due Dr. Harold A. Nash and Dr. Brooks D. Church of
North Star Research and Development Institute and Mr. Willard Brosz of
the Green Giant Company, who were responsible for conducting this
program.
Several others are acknowledged for having made significant con-
tributions including, Dr. Eugene E. Erickson for his many suggestions
and guidance through the engineering phases of this study, and
Mr. Dale Bergstedt for his efforts in securing many items of equipment
and for coordinating the efforts of the Green Giant Company, North
Star, and commercial people who loaned or rented some of the equip-
ment. Mr. Clarence Sprague of Green Giant was most helpful through
his many suggestions and ideas in the construction and maintenance
of the pilot-plant facilities. The cooperation, timely help, and
friendliness expressed by all members of the Agricultural Production
Department of the Green Giant Company, Glencoe, Minnesota, are
especially acknowledged. Special appreciation is due Mr. Donald
Hartung for his continued, responsible vigilance of the pilot-plant
facilities throughout the study.
The support of the project by the Environmental Protection Agency, and
the help provided by Mr. Kenneth Dostal, the Grant Project Office, and
Mr. H. George Keeler are acknowledged with sincere thanks.
73
-------
SECTION IX
REFERENCES
1. Church, B.D., and Nash, H.A., Use of Fungi Imperfecti in Waste
Controlf Water Pollution Control Research Report 12060 EHT
(July 1970).
2. Standard Methods for the Examination of Water and Wastewater^
American Public Health Association, New York, 12th Ed (1965).
3. Conway, E.J., Mi orodiffusion Analysis and Volumetric Errorf
McMillan, New York, 4th Ed, p. 98 (1958).
4. Johnson, M.J., "Determination of Nitrogen by the Micro-Kjeldahl
Method", J. Biol. Chem.f 137, 575 (1941).
5. Fiske, C.H., and Subbarow, Y., "The Determination of Inorganic
Phosphates in Whole Blood, Plasma, or Serum", J. Biol. Chern., 66,
375 (1925).
6. Eckenfelder, W.W., and O'Conner, D.J., Biological Waste Treatment,
Pergamon Press, Oxford, p. 21 (1961).
7. Pipes, W.O., "Types of Sludge which Separate Poorly", J. Water
Pollution Control Fed., 41 f 714 (1969),
75
-------
SECTION X
LIST OF PATENTS AND PUBLICATIONS
Church, B.D., Nash, H.A., Erickson, E.E., and Brosz, W., Continuous
Treatment of Corn and Pea Processing Wasteuater with Fungi Imperfeatit
Proceedings of Second National Symposium on Food Processing Waste, Denver,
Colorado, March 23-26,1971, Pacific Northwest Water Laboratory of the EPA
and National Canners Association.
Church, B.D., Nash, H.A., and Brosz, W., "Use of Fungi Imperfeoti in
Treating Food Processing Wastes", Developments in Industrial Micro-
biology, Soc. Indust. Microbiol., A.I.B.S., Washington, D.C. (in press).
77
-------
Number
Subject Field &. Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
The Green Giant Company, Le Sueur, Minnesota 56058 and North
Development Institute, Minneapolis, Minnesota 55406
and
Title
PILOT-PLANT INSTALLATION FOR FUNGAL TREATMENT OF VEGETABLE CANNING WASTES,
]Q Authors)
Nash, Harold A.
Brosz, Willard
16
21
Project Designation
12060
EDZ
Note
22
Citation
23
Descriptors (Starred First)
*Fungi Imperfecti, *Vegetable Waste, *Pilot-Plant Installation, Process Costs
25
Identifiers (Starred First)
*Fungal Degradation, Pilot-Plant Installation
27
Abstract
The use of the imperfect fungus, Triohodermo. viridet to treat com and pea wastes
has been tested in continuous fermentation at the 10,000-gallon scale. Both a pool
unit and an oxidation ditch were tested. The pH was 3.7, and ammonium ion and phosphate
were added. The average residence time was 20 hours. An aerated lagoon was also
operated to compare with the two fungal systems.
In the fungal systems, about 96 percent removal of BODs, 88 percent removal of
COD, and 93 percent removal of TOG was achieved on corn canning wastes. Performance
on pea canning wastes was somewhat less. Essentially zero levels of ammonia
nitrogen and inorganic phosphate could be attained in the effluent stream.
Fungal yields, on a dry-weight basis, were about 50 percent of the BODs of
the feed, and the protein content of the dry mycelium was about 50 percent.
Costs are estimated at 4.9 cents per pound of BOD5. Sale of mycelium as
feed could decrease this to 3.1 cents. Operation on a year-around basis with
sale of the mycelium would decrease costs to about 1.1 cents per pound of BODs.
Direct feeding of the mycelium without drying could further reduce the net cost
to about 0.8 cent per pound of
Abstractor
. Brooks D. Church
Institution
North Star Research and Development Institute
WR:I02 (REV JULV
WRSIC
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
------- |