EPA-660/2-75-002
FEBRUARY 1975
Environmental Protection Technology Series
Pilot Scale Treatment of Wine Stillage
National Environmental Research
Office of Research and Development
U.S. Environmental Protection Agency
Corvallis, Oregon 97330
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY STUDIES series. This series describes research
performed to develop and demonstrate instrumentation, equipment
and methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the Office of Research and
Development, 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 mention
of trade names or commercial products constitute endorsement or
recommendation for use.
-------
EPA-660/2-75-002
February 1975
PILOT SCALE TREATMENT OF
WINE STILLAGE
by
E. D. Schroeder
Department of Civil Engineering
University of California
Davis, California 95616
Grant No. 12060 HPC
Program Element 1BB037
ROAP/TASK No. 21 BAC/010
Project Officer
Max W. Cochrane
Pacific Northwest Environmental Research Laboratory
National Environmental Research Center
Con/all is, Oregon 97330
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS, OREGON 97330
-------
ABSTRACT
y
Pilot and laboratory scale studies were run on aerobic and anaerobic
biological treatment of winery stillage over a two year period. The
pilot scale studies included work with aerobic lagoons and anaerobic
packed towers. Laboratory systems studied were aerobic reactors
without recycle and batch fed anaerobic processes. Because suspended
solids removal proved to be a key factor in successful biological
treatment, centrifugation, detartration, coagulation and flocculation,
and combinations of these methods were included in the studies.
Centrifugation proved to be the best method of removing solids prior
to biological treatment. Solids removal in combination with an
aerobic treatment process can be expected to produce final filtrate
chemical oxygen demands of about 700 mg/L and a final filtrate BOD
of about 75 mg/L. Anaerobic processes studied did not operate well
but produced effluents with chemical oxygen demands of the order of
4000 mg/L.
This report was submitted in fulfillment of Grant No. 12060 HPC by
the California Department of Agriculture, Wine Advisory Board, 717
Market Street, Sari Francisco, CA 94103 under the partial sponsorship
of the Environmental Protection Agency. Work was completed as of
July 30, 1973.
11
-------
CONTENTS
Page
Abstract ii
List of Figures iv
List of Tables v
Acknowledgements vii
Sections
I Conclusions 1
II Recommendations 3
III Introduction 4
IV Experimental Systems and Methods 16
V Experimental Studies and Results 32
VI Discussion 72
VII References 81
VIII Publications 84
IX Glossary 85
X Appendices 86
m
-------
FIGURES
No. Page
1. Wine Process Schematic 9
2. Cascade Reactor System Schematic 15
3. Pilot System Schematic 18
4. Aerobic Treatment Unit 20
5. Anaerobic Packed Bed Reactor 21
6. Typical Raw Still age Settling Curve 41
7. Stillage Feed Concentration Versus Hydraulic
Detention Time for Laboratory Reactors 42
8. COD Versus Detention Time for Reactor 1 47
9. COD Versus Detention Time for Reactor 2 in Series 48
10. Specific Growth Rate Versus Unit Substrate
Removal Rate 50
11. Settling Curves for Purifloc C-41 Amoco
Cationic Mixture 56
12. Effect of Coagulant Addition on Supernatant Liquor 57
13. Cake Moisture Vs. Centrifuge Run Time 58
14. Suspended Solids Removal Versus Centrifuge Run Time 59
15. Supernatant Liquor COD Versus Centrifuge Run Time 60
16. Cake Moisture Vs. Centrifuge Run Time 62
17. Suspended Solids Removal Versus Centrifuge Run Time 63
18. Supernatant Liquor Versus Centrifuge Run Time 64
19. Supernatant Liquor pH and Filtrate COD Versus CaClo
and Ca(OH)2 65
20. Settling of Detartrated Raw Stillage 55
21. Supernatant COD Versus Coagulant Concentration 68
22. pH and COD Versus CaCl2 and Ca(OH)2 Concentration 70
23. Suspended Solids and COD Concentration of Detartrated
Settled Supernatant Liquor 71
24. Proposed Treatment System 74
25. Estimated Total Cost of Centrifuges, Aerators
and Basins 79
-------
TABLES
No. Page
1. Pomace Still age Characteristics 6
2. Polyelectrolytes Used in Coagulation Experiments 25
3. COD Analysis of Fresno Samples (filtered) 36
4. COD Analysis of Fresno Samples (unfiltered) 38
5. Mixed Liquor Suspended Solids Concentration 45
6. Operational Variables for First Laboratory
Reactor in Series 46
7. Operational Variables for Second Laboratory Reactor 46
A-l. 1971 Pilot Plant Influent COD Data 87
A-2. 1971 Activated Sludge Pilot Plant 88
A-3. 1971 Unfiltered Anaerobic Packed Bed COD Data , 89
A-4. 1971 Anaerobic Packed Bed Filtrate COD Data 90
A-5. 1971 Temperature and pH Data for Aerobic and
Anaerobic Systems 91
B-l. 1972 Aerobic Pilot Plant Data 92
B-2. September 24, 1972 Anaerobic Packed Bed Number 1
Data 92
B-3. September 24, 1972 Anaerobic Packed Bed Number 2
Data 93
B-4. September 12, 1972 Anaerobic Packed Bed Number 3
Data - 93
B-5. September 24, 1972 Anaerobic Packed Bed Number 3
Data 93
B-6. 1972 Raw Still age Settling Data 94
B-7. 1972 Activated Sludge Mixed Liquor Settling Data 95
C-l. Laboratory Reactor 1 At Gc = 4.67 Days 98
-------
TABLES (cont.)
No. Page
C-2. Laboratory Reactor 2 At Gc = 4.67 Days 99
C-3. Laboratory Reactor 1 At 0C = 3.58 Days 100
C-4. Laboratory Reactor 2 At 0C = 3.58 Days 101
C-5. Laboratory Reactor 1 At 8C = 2.75 Days 102
C-6. Laboratory Reactor 2 At Qr = 2.75 Days 103
\+
C-7. Laboratory Reactor 1 At 6C = 2.42 Days 104
C-8. Laboratory Reactor 2 At 0C = 2-42 Days 105
C-9. Laboratory Reactor 1 At 0C = 1.87 Days 106
C-10. Laboratory Reactor 2 At 0C = 1.87 Days 107
C-ll. Laboratory Reactor 1 At ec = 1.41 Days 108
C-12. Laboratory Reactor 2 At 0C = 1.41 Days 109
C-13. Laboratory Reactor 1 At 0C = 1.09 Days 110
C-14. Laboratory Reactor 2 At 0C = 1.09 Days 111
D-l. Batch Settling Test Results for 1/3, 2/3 Mixture
of Purifloc C-41 and Amoco Cationic Coagulant Aids 112
D-3. Effect of Centrifugation on Cake Moisture Centrate
Suspended Solids and Centrate COD 113
D-4. Effect of Polyelectrolyte Addition on Centrate
Characteristics at 2400 RPM 113
D-5. Effect of Detartration on Supernatant COD 114
D-6. Settling of Detartrated Still age Using Purifloc
A-23 115
D-7. Coagulation of Centrate with Nalco 610 Polyelec-
trolyte and Bentonite 117
D-8. Detartration of Centrate 118
D-9. Coagulation of Detartrated Centrate with
Purifloc A-23 118
-------
ACKNOWLEDGEMENTS
Major contributors to this study were D. J. Reardon and W. H. Hovey
of the Department of Civil Engineering, and Professor A. P. Jackman
and Robert Matteoli of the Department of Chemical Engineering. The
participation of Professor Jackman was particularly appreciated.
D. C. Turrentine and John Lockett of the California Wine Advisory
Board were extremely helpful throughout the project and this was
greatly appreciated. Hugh Cook of the Wine Institute provided a
great deal of help with communications and background information.
Charles Crawford of the Gallo Winery provided support and advice
throughout the study. Without his help no progress would have been
made. Particular thanks also are in order for Ted Yamada, Fresno
Winery manager in 1971 and Bob SI ayton, Fresno Winery manager in 1972,
for their continual consideration and patience.
Kenneth Dostal of the National Environmental Research Laboratory,
Corvallis, Oregon was of considerable help in the projects development
and demonstrated continual interest over the two year period. The
project officers have continually supported the project and this was
greatly appreciated.
Vll
-------
SECTION I
CONCLUSIONS
1. Biological treatment of California Winery stillage is possible.
Aerobic (aerated lagoon) processes are the most feasible systems
for treatment and these should be designed as two stage cascades
in order to minimize nitrogen requirements.
2. Pretreatment to reduce suspended solids concentrations below
2000 mg/L is necessary for successful biological treatment
process operation. Foaming and oxygen transfer problems are
uncontrollable at higher concentrations. Centrifugation is
the most suitable solids removal method because sludge
concentrations produced are of the order of ten percent by
weight or greater. Stillage suspended solids concentrations
are of the order of two percent by weight and therefore sludge
concentrations below ten percent result in an unacceptably
large sludge volume.
3. Stillage acidity is not a major process control problem.
Because the acidity is due to biodegradable organic acids,
the process pH can be controlled by matching the loading
rate to the organic removal rate without chemical addition.
4. The maximum organic removal rate was found to be 12.5 grams
COD removed/liter-day, and the corresponding loading rate was
14.1 grams COD/1iter-day. At higher loading rates, the pH
and the organic removal rate decreased rapidly. Effluent
quality at the maximum loading rate is approximately 2300 mg/L
1
-------
COD in the settled supernatant, 650 mg/L COD in the filtrate
and 75 mg/L BOD5 in the filtrate.
Nutrient addition can be restricted to nitrogen. Very little
nitrogen is available in the raw still age and the stoichiometric
requirement is approximately 2 grams N per liter. Nitrogen
addition can be minimized by adding it only to the second of
two bio-oxidation units operating in series. Results of the
1972 pilot plant studies lead to the conclusion that only about
500 mg/L N must be added at that point. Nitrogen should be
added as NH,NQ~ to avoid greatly increasing the sodium (from
adding NaN03) or chloride (from adding Nll^Cl) concentrations
of the still age.
Foam and fly control in the biological treatment processes is a
major problem. A stiff, hard to break foam is generated during
aeration which provides an excellent media for flies to deposit
eggs. Foaming was considerably less at longer residence times
in the laboratory studies (three days or greater), but was a
continual problem in the field studies which had a residence
time of three days. A method of foam control is necessary if
treatment is to be successful.
Success of pilot scale anaerobic treatment studies was limited
because of poor temperature control. Laboratory study results
indicate that effluent quality will be lower than that from
aerobic processes. Start up problems were not noted during the
second year of the field studies, but the pilot processes never
produced effluent filtrate COD concentrations below 3900 mg/L.
-------
SECTION II
RECOMMENDATIONS
Wine still age treatment will be expensive from both capital
expenditure and operational cost points of view. For this
reason treatment should be considered as an alternative
available when land disposal is impossible. If a decision is
made to treat still age at a winery, lack of experience in
still age centrifugation and foam and fly control will present
problems. Additional research and development effort will be
necessary in these areas.
-------
SECTION III
INTRODUCTION
Wastewaters from California wineries include effluents from
processing operations, tank cleaning and distillation of beverage
brandy and fortifying spirits. Of these discharges, the most
difficult treatment problem is associated with the wastev/ater from
the distillation process. This wastewater, normally called
still age or still slops, consists of the nonvolatile material from
the bottom plates of continuous stills or the residue remaining in
batch stills. Still age production varies from winery to winery,
depending on the quantity of sweet wines and brandy produced, and
on the type of still used. Volume of stillage produced per ton of
grapes processed is not a useful parameter because not all wineries
produce distilled products and those that do vary considerably in
the amount of distilling material needed. A medium size operation
will produce around 150,000 liters/day (40,000 gallon/day) and a
large installation may produce as much as 2,300,000 liters/day
(600,000 gallon/day). Nearly 90% of California's winery distillation
operations occur from late August to early November. Thus most of
the waste is generated during the 45 to 75 day period in which
crushing occurs^ Virtually all of the California wineries which
produce distilled products are located in the San Joaquin Valley, and
the problem is, in practical terms, limited to this region.
Currently most wineries dispose of stillage by discharging into
municipal sewerage systems, treatment in lagoons or by land disposal
through intermittent irrigation. Municipal systems receiving stillage
have usually experienced operational problems due to overloading, and
-------
odor problems are often associated with conventionally designed
lagoons. Intermittent irrigation has proven to be the most
satisfactory method of disposal. Pretreatment is not required,
nuisance problems are less than those associated with other
treatment methods and the method is particularly ameniable to
seasonal operations. Many wineries are located in areas where
land is either unsuitable for irrigation or becoming less
available. Thus these wineries in particular, and the industry
in general, are interested in developing alternative treatment
methods.
Fermentation residues are the major source of distilling material
during the processing season. The first phase of the fermentation
process produces pomace material such as settleable skins, seeds and
pulp, and the second fermentation process produces lees, which are
yeast and solids coagulated with bentonite and settled. When liquid
from pomace or lees materials are used in a distillation process the
waste is called pomace or lees still age, respectively.
Still age characteristics vary considerably with the source of the
distilling material (lees, pomace or wine), the operation of the
winery and the type of still used. A general characterization would
be that still age is very high in COD, BOD, suspended solids and
acidity however. Typical values reported for pomace stillage are
given in Table 1.
Several significant treatment process design and operational problems
can be foreseen from waste characteristics discussed up to this point.
The high organic concentrations restrict the process choice to
anaerobic systems or aerobic systems with a high oxygen transfer
-------
Table 1. POMACE STILLAGE CHARACTERISTICS ]'2»3»4
Characteristic Range Reported
pH 3.5 - 5.0
Acidity, mg/L as 1,200 - 3,800
Total Solids, mg/L 13,000 - 30,000
Suspended Solids, mg/L 14,000 - 18,000
Volatile Solids, mg/L 10,000 - 27,000
Total BOD5, mg/L 2,400 - 17,840
Total COD, mg/L 34,000 - 53,000
Filtrate COD (0.45 Micron), mg/L 19,000 - 22,000
Total Nitrogen, mg/L as N 150 - 330
NH3 - N mg/L as N 2-4
Total Phosphorous, mg/L as P 1,211 - 1,310
Temperature (at the still) 150°F
-------
rate (in terms of mass/time), and the extremely high suspended solids
concentrations make satisfactory solids removal by conventionally
used techniques very difficult. Acidity and pH values associated
with still age force the use of some form of pH control in the treatment
, process. Nearly all of the nitrogen in the waste is in the organic
form and therefore is available for use in a treatment process only
as fast as the organic nitrogen containing compounds are broken down.
In addition, the total nitrogen in the wastewater falls far short
of that needed in aerobic biological treatment process. Finally the
seasonal nature of current distillation operations requires that a
treatment system has a very short start up time.
OBJECTIVES
i
The primary objectives of this project were waste characterization,
pilot and laboratory scale investigation of aerobic and anaerobic
biological treatment of still age and development of process design
criteria. Preliminary evaluation of treatment costs, alternative
treatment methods, such as direct pomace fermentation and effect of
grape variety on treatment system operation were also included in the
objectives.
BACKGROUND
Distilling material produced during the crushing season is made up of
lees, pomace, stems and leaves and washwater. Pomace includes yeast,
pulp, skins and seeds separated from the wine after the first
fermentation. This material is usually washed out of the fermentation
tank, dewatered and pressed. Wine from the press is recovered and
the pressed pomace is mixed with stems, leaves and water in the
7
-------
disintegrater and scalper. Liquid from this mixture, commonly called
pomace distilling material, is then held for distillation. Additional
fermentation of residual sugar takes place during storage. Lees
material results from precipitation of solids with bentonite at the
end of the second fermentation step. In addition to bentonite, lees
would be expected to contain yeast and residual solids from the first
fermentation. Water is added to lees as they are washed out of
fermentation tank. Solids removed during final clarification of
the wine before bottling are usually mixed with the lees. These
solids may include coagulant aids such as bentonite, gelatine or
casein.
Distilling material production resulting from red wines is slightly
different from that for white wines. White wine production requires
removal of the skins at the crusher along with the stems to prevent
coloring during fermentation. This procedure has little, if any,
effect on the distilling material characteristics. A schematic of
the stillage production process is shown in Figure 1.
Alcohol collected from the distillation process is used to stop
fermentation during the production of sweet (dessert wines).
Addition of the alcohol takes place during the second fermentation
step.
CURRENT METHODS OF STILLAGE DISPOSAL
Current methods of still age treatment include lagooning, anaerobic
treatment and interim'ttant irrigation. Anaerobic treatment has been
5 6
used in South Africa^' with wineries which operate most of the year.
-------
GRAPES
i
WASH
WATER
CRUSHER
FIRST PHASE
FERMENTATION
(3-6DAYS)D
SECOND PHASE
FERMENTATION
(2-6 WEEKS )
*» STEMS i
_^^_1M^M Fk ^ !*• 1 fc 1 T F* rt Fl ft T
^| DESINTEGRAT
WASH
WATER PRESSED POMACE
1
OR
1
f DEWATERING POMACE
SCREENS PRESS
WASH
WATER to
I
i +
1 LEES
T
J SCALPER
1
SOLIC
i i
DISTILLING
MATERIAL
STORAGE ALCOHOL
FINING
BLENDING
AGING
FILTERING
_J f
STILL
1
I
« BOTTLING
STIL
•^—
LAGE
FIGURE 1. WINE PROCESS SCHEMATIC
-------
Systems used are essentially anaerobic contact processes' i.e. the
cell and hydraulic residence times are different. Hydraulic residence
times are of the order of 7 to 8 days and effluent COD concentrations
are of the order of 500 mg/L. No anaerobic processes are being
2
intentionally operated in the California area, although Pearson et. al.
o
and Chadwick and Schroeder reported on laboratory scale anaerobic
treatment studies. A major problem with anaerobic processes is the
slow start up rates which restrict their use for treating seasonal
wastes.
Lagoons are used by a number of wineries in the San Joaquin Valley.
Odor complaints have been a general problem associated with these
systems. Unfortunately design loadings associated with the production
of odors are not known .
Intermittant irrigation has proven to be the simplest and least
o
expensive method of stillage disposal . Maximum loading rates are
100,000 gallons per acre per week. Operation is on a batch basis.
Furrows are filled and a period is allowed for evaporation and
percolation. Dry solids residue is then removed or disced into the
soil and the field can be reused. Nuisance problems with this
procedure are not great if soil conditions are appropriate and if the
disposal area is separated from residential or commercial areas.
There has been some concern about potential damage to ground water
Q
quality from irrigation , but information is not available which
would allow evaluation of the problem. York has recently reported
on studies of soil core samples in fields when intermittent irrigation
has been used. His tentative conclusion is that salt transmission is
not a significant problem. Tile drainage would undoubtably be a
solution if a threat to ground water is demonstrated in later studies.
10
-------
Trickling filters 1>12 and activated sludge processes13'14 have been
used to some extent for the treatment of still age, but with little
success. Rates of oxygen demand have invariably been larger than
the oxygen transfer capacities of trickling filters used.
Filamentous growths have been the major problem in the operation of
activated sludge processes. The University of California at Davis
has been involved in the study of still age disposal methods since
the revival of the California wine industry in 1933. fluch of the
work has been by Vaughn and his coworkers11'15'16. Studies have
included both still age characterization and methods of still age
treatment.
CONCEPTUAL APPROACH
Still age treatment system design and operation must take into
consideration the low pH, and low nutrient concentrations and the
high acidity, organic content, solids content and temperature of
the wastewater. Each of these factors places constraints on the
system. For example neutralization with lime results in excessive
chemical costs and excessive sludge production due to the low pH
and high acidity. The high organic content of still age results in
high cell production rates. In addition, a significant fraction
of the material making up the COD can be expected to consist of
lignins, cellulose and other difficult to degrade organics.
Suspended solids in the still age make up approximately one half of
the COD, but consist primarily of pulp, seeds, skins, stems and
yeast cells which pass a 0.24 cm (3/32") dewatering screen and
which would be expected to be difficult to bio-oxidize. In
addition, the high suspended solids concentrations will decrease
potential oxygen transfer rates. Still age is nitrogen limited, but
11
-------
the extent of the limitation is difficult to determine. Nearly all
of the nitrogen is in the organic state and therefore bio-oxidation
must take place before it becomes available for bacterial growth.
The extent of nitrogen limitation is therefore dependent on the
quantity of nitrogen tied to the nondegradable organics. If all
of the organic nitrogen is unavailable nitrogen addition necessary
may be as high as 2 grams/liter. Limitations are also imposed by
the temperature of still age. The 66°C temperature reported in
Table 1 is immediately following the heat exchanger at the still
and the treatment plant. Temperatures in this range would be ideal
for anaerobic treatment processes although some additional heat
would probably have to be added to maintain thermophillic conditions.
Filamentous cultures seem to predominate when high temperatures are
(17) maintained in aerobic processes, however.
The constraints on biological waste treatment of still age imposed
by the still age characteristics can be summarized by saying that a
large fraction of the suspended solids must be removed prior to
aerobic treatment, pH must be maintained near neutral within the
biological processes and, nutrients will have to be added to
aerobic treatment processes to satisfy stoichionetric requirements
for growth. Temperature control will have to be considered also,
but this is not a major problem in full scale systems. Heating of
anaerobic processes is standard practice and cooling will occur
during solids removal and aeration in aerobic processes.
A major constraint on the process choice is imposed by the seasonal
nature of the wastewater production. Product quality control prevents
long term storage of the distilling material (19) and thus the problem
is inherrently seasonal. Stillage production begins within a few days
12
-------
of the start of crushing and reaches maximum flow rates within a week.
Therefore, the waste treatment process chosen must have a short start
up period.
Strength of the waste is also an important consideration. The high
soluble COD concentrations, most of which is biodegradable (3,5,6)
places constraints on both aerobic and anaerobic processes. Aerobic
process design will be limited by the ability to transfer oxygen,
while anaerobic systems will be limited by the rate of conversion of
organic acids to methane. In either case, the constraints will be
represented by minimum hydraulic and mean cell residence time values.
The initial research plan was based on the considerations discussed
above. Activated sludge was chosen as the pilot aerobic treatment
system and packed bed anaerobic treatment (often called the anaerobic
filter) was chosen as the anaerobic process for pilot scale
investigation. Determination of acceptable loading rates, possible
effluent quality and characteristic operating problems were the
immediate objectives of the studies. The choice of the packed bed
anaerobic treatment process was based on the need for short start up
times. Because the cells have a much longer residence time than the
wastewater in this system, cell concentrations become very high.
Suspended solids retained in the units with the cells can serve as a
partial food source during the off season and it was felt that a
satisfactorily short start up period would possibly occur during the
second season of operation. Because of the high solids content of
the wastewater plugging of the packed bed was a distinct possibility.
For this reason a sedimentation tank was placed ahead of the packed
bed as well as ahead of the activated sludge processes. Sedimentation
tank sizing was based on Chadwick and Schroeder's recommendation of
13
-------
maximum overflow rates of 2034 liter/square meter/day (50 gallons/
square foot/day) and the maximum expected flow rate.
Initial design of the pilot activated sludge process utilized a
cascade reactor concept (Figure 2). The cascade was to be a series
of continuous flow stirred tank reactors with the option of settling
and cell recycle between each stage. Growth rates would be very high
in the first reactor and progressively decrease to the final reactor.
Settleability of the culture was expected to improve through the
cascade with a highly dispersed culture in the first stage and a
progressively more flocculant culture developing as the growth rate
decreased. Discussions with Kenneth Postal of the National
Environmental Research Center of the Environmental Protection
Agency (Corvallis, Oregon) in March, 1971 resulted in altering the
conceptual design to a single large reactor and a two tank cascade
operated in parallel.
The program of study involved construction of the pilot plant units
during the summer of 1971, and operation of the pilot plant system
during the 1971 processing season. Observations and data from the
1971 season was used to set up laboratory studies during fall, winter
and summer of 1971-1972. These studies were then used to set
operating criteria for the pilot plants during the 1972 processing
season.
-------
RAW
STILLAGE
REACTOR
r L
REACTOR
WASTE WASTE
f
WASTE
FIGURE 2 . CASCADE REACTOR SYSTEM SCHEMATIC
-------
SECTION IV
EXPERIMENTAL SYSTEMS AND METHODS
PILOT PLANTS
Pilot plant studies were carried out at the Gallo Winery in Fresno,
California. The winery furnished a concrete pad, electricity and
water at the experimental side and also provided a shed for storage
of chemicals and equipment. Help with mechanical and electrical
problems which occurred from time to time was also furnished by the
winery. The concrete pad was directly over the pipe carrying
still age to the fields used for intermittant irrigation and at a
point approximately 300 meters from the still. Stillage for the
studies was supplied from a tap in the pipe.
Stillage was pumped from the sewer into a hopper bottomed holding
7 9
tank having a surface area of 1.9 meter (20 ft. ) and a volume of
3 3
1.9 meter (66.7 ft. ). Design residence time in this holding tank
o
was approximately six hours. Previous studies had led to the
conclusion that solids removal would not be great in this tank and
therefore identical tanks which acted as both holding and
sedimentation tanks were placed in front of each group of biological
treatment orocesses.
Activated SIudge Processes
Two continuous flow, stirred tank activated sludge systems were
operated in parallel. One system consisted of a single aeration
basin with an attached sedimentation tank and the second system
16
-------
consisted of two identical aeration basin-sedimentation tank
combinations in series (Figure 3). Total aeration tank volume of
the two systems was equal. Dimensions of the large aeration basin
were 0.915 meter (3 ft.) wide, 1.83 meter (6 ft.) long, and 1.33
meter high. Liquid depth was 1.36 meter (4.45 ft.). The smaller
units differed in that they were only 0.915 (3 ft.) in length.
Each aeration basin had an attached sedimentation basin with
dimensions of 0.915 meter by 0.61 meter (2 ft.). The units were
hopper bottomed and had a total volume of 0.57 meter (20 ft/).
Solids were removed at the tank bottom and pumped back into the
aeration basin. Each sedimentation tank was divided into two
sections by a partition which allowed operation at two different
overflow rates for any given flow rate.
Design hydraulic residence time of the activated sludge systems
was 12 hours. This corresponds to a flow rate of 190 liters/hour.
Design overflow rate in the holding-sedimentation tank was therefore
2400 liters/meter2-day (59 gallons/ft.2-day) and the design overflow
2
rate in the secondary clarifiers was either 8170 liters/meter"-day
(200 gallons/ft.2-day) or 16340 Iiters/meter2-day (400 gallons/
ft.2-day).
Mixing and aeration of the activated sludge units were provided by
0.61 meter (2 ft.) diameter flat bladed turbines turning at 70 rnn.
The large aeration basin had two turbines and the small aeration
basins had one. Power for the turbines was provided by 1490 watt
(2 hp) Baldor motors (Baldor Manufacturing Company, Fort Smith,
Arkansas) operating at 1725 rpm.
17
-------
SMALL
AEROBIC
UNIT
LARGE
AEROBIC
UNIT
SMALL
AEROBIC
UNIT
FIGURE 3 PILOT SYSTEM SCHEMATIC
-------
A sketch of the activated sludge units is given in Figure 4.
Anaerobic Packed Beds
The anaerobic packed beds were designed on the basis of a nominal
maximum loading of 9.6 grams COD/liter/day (0.6 Ib. COD/ft.3/day)
and an influent COD concentration of 20,000 mg/L (taken from
Chadwick and Schroeder's (3) settled stillage COD values). Three
0.915 meter (3 ft.) diameter and 2.44 meter (8 ft.) deep steel
tanks were constructed to house the anaerobic processes. An influent
distribution section was provided in the lower 0.46 meter (18 inches)
of the tank and a 0.46 meter freeboard was provided at the top. The
1.83 meter section remaining was packed with 5 cm (2 inch) Douglas
Fir bark chips. Packing was contained by expanded metal grates
placed at the top and bottom of the units. Sampling ports were
placed at 0.3 meter (1 ft.) intervals in the packed volume as shown
in Figure 5. For the anaerobic packed apparent (unpacked) volume
the flow rate corresponding to the maximum loading rate was 240
liters/hr. (63.6 gallons/hr.) and the apparent hydraulic residence
time was 5 hrs. Three identical units were constructed and operated
in parallel. Design flow rates were 240 liters/meter, 120 liters/
hour and 60 liters/hour.
Pumps and Flow Control
Pumps used were Jabsco model B3-M6 (Jabsco Pump Company, Costa Mesa,
California) rated at 38 liters/minute (10 gpm) at zero head. Flow
rate control v/as provided by use of cam timers. For example, to
maintain an average flow of 4540 liters/day for the activated sludge
systems timers with a five minute cycle were chosen and operated in
19
-------
2 HP MOTOR
INFLUENT
SLUDGE
FIGURE 4 . AEROBIC TREATMENT UNIT
20
-------
OVERFLOW
INLET
MANIFOLD
PACKING
MEDIA
GRATE
SAMPLE
PORTS
$30 CM
GRATE
FIGURE 5. ANAEROBIC PACKED BED REACTOR
21
-------
the on position for 25 seconds per cycle. Design flow rates reported
here are based on pump characteristic curves, measured head, and the
timing cycles used. Other flow rates reported were measured on a
volumetric basis over several cam timer cycles.
Pomace Fermenters
Pomace fermentation is conventionally carried out after mixing of
the pomace with stems in the disintegrator and washing with water
(see Figure 1). The pomace is partially fermented prior to this
operation, but residual sugars and sugars washed off of the stems
are converted to alcohol at this point. Alcohol concentration is
low (1 to 4%) but economic recovery is possible. Because most of
the potential alcohol is in the pomace, the possibility exists for
fermenting the pomace without the stems and wastewater and greatly
decreasing the wastewater flow rate. Temperature control is the
primary concern because the material being fermented would be pressed
pomace which is similar to a filter cake. The purpose of the
experiments in this portion of the study was to determine the amount
of heat generated during fermentation with the objective of developing
a proposed method of temperature control.
Pilot scale pomace fermenters were designed and constructed during
the second year of the project. Because the primary problem was
expected to be temperature increases within the fermenting pomace,
the units were designed to minimize heat exchange and allow
semi continuous temperature monitoring at four points. Three
identical fermenters were constructed. Each cylindrical plexiglass
unit had an inside diameter of 28 cm, an insulation space between
the inner and outer walls of 1.8 cm, and a length of 61 cm.
22
-------
Thermometers were inserted three points along the length of the
cylinder. The thermometer position could be adjusted radially
allowing development of complete temperature profiles.
LABORATORY SYSTEMS
Laboratory scale work included aerobic biological treatment, anaerobic
biological treatment, and solids removal studies. Stillage used in
all of the laboratory studies was obtained at the Gallo Winery in
Fresno, transported and stored on the Davis Campus at -30°C. The
still age used in the solids removal studies was dewatered by flash
evaporation at the winery and reconstituted after thawing, prior to
use. Tap water was used in reconstituting the still age and the
dilution factor used was such that the filtrate COD was 13000 mg/L.
Changes in still age characteristics during storage were not detected.
Activated Sludge
Laboratory aerobic biological treatment systems consisted of tv/o
3.5 liter plexiglass stirred reactors in series without cell recycles.
Diffused air v/as the source of oxygen and mixing energy. Still age
fed to the units was the supernatant liquor from thawed 20 liter
aliquots which had been allowed to settle for 24 hours. A variable
speed Masterflex tubing pump, model V-13 (Cole-Parmer Co., Chicago)
controlled by a cam timer was used to feed the still age to the
system. Temperature was maintained at 23°C by housing the entire
experimental system in a controlled temperature room.
The pH of stillage fed to the aerobic units was not adjusted, but
nitrogen (3.6 g/L NH.C1) was added. Hydraulic, and therefore mean
23
-------
cell residence times used, ranged from 1.1 days to 4.7 days in
each unit. Thus, total residence time in the two reactor series
varied between 2.2 and 9.4 days. System pH, COD and mixed
liquor suspended solids concentration was monitored until steady
state conditions were achieved at each residence time studied.
These conditions were maintained for one to two weeks of operation.
Anaerobic Treatment
The laboratory anaerobic treatment studies were conducted using
1.5 liter continuously mixed batch reactors. Operating temperature
of the units was 57°C (135°F). This temperature was chosen on
the basis of still age being available at a temperature of 66°C at
the still. Operation was on a daily fill and draw basis. Three
residence times, 15 days, 30 days and 60 days were used. Settled
still age was used in these experiments which had a COD value of
15,500 mg/L. Gas production in the anaerobic systems was measured
by liquid displacement.
Solids Removal Studies
Solids concentration and removal methods considered for pomace
still age were: a) coagulation with polyelectrolytes, flocculation
and sedimentation, b) centrifugation, c) centrifugation with
polyelectrolyte addition, and d) detartration with polyelectrolyte
addition, flocculation and sedimentation. Plain sedimentation and
o
dissolved air flotation were considered by Chadwick and were not
included in these studies. All of the solids removal studies were
conducted at room temperature (approximately 21°C).
24
-------
Coagulation -
Coagulants used were limited to polyelectrolytes because of the low
pH and high acidity of the stillage. Use of cationic, anionic and
nonionic polymers was investigated. Polymers used are listed in
Table 2.
TABLE 2
POLYELECTROLYTES USED IN COAGULATION EXPERIMENTS
CATIONIC ANIONIC NONIONIC
Purifloc C-31 Purifloc A-23 Purifloc N-12
Purifloc C-41 Amoco Anionic Amoco Nonionic
Amoco Cationic Separan MC 200 Separan MGL
Nalco 610 Nalco 607
Nalco 610-HD Bentonite Nalco 634
Nalco 671
Maico 676
Coagulation tests were run using Standard Jar Test Apparatus and
19
methods except for pH adjustment. Polyelectrolyte concentrations
used were in the range of 10 to 200 mg/L. In each case an appropriate
quantity of polyelectrolyte was added to 500 ml of reconstituted
still age. The mixture was then stirred at 100 rpm for 10 minutes and
fTocculated for 2 minutes at 20 rpm. After flocculation the mixture
was poured into a 500 ml graduated cylinder and allowed to settle.
Clear water interface height and supernatant liquor COD were measured
as functions of time.
25
-------
Centrifugation -
Variables considered in the centrifugation studies included rotational
speed and time of centrifugation. All experiments were conducted using
75 ml samples and an IEC model UV centrifuge. Rotational speeds and
run times used were 1200, 1800, 2400 and 3000 rpm, and 1, 2, 4, 8 and
15 minutes, respectively. Cake moisture (in %), supernatant liquor
COD and supernatant liquor suspended solids concentration were recorded
as functions rotational speed and run time.
Comparison of centrifuges is made by using an index termed relative
22
centrifugal force (RCF) :
RCF = 0.00001117 r N2 (1)
where:
r = radius in cm
M = speed of rotations in RPM.
Proper test data includes the size of tubes used, time of centrifugation,
and RCF values at the tip of the tube and free surface of the liquid.
The usefulness of RCF is that when data is to be compared or developed
using different apparatus the depth of liquid at a given angular
velocity which will correlate with previous results can be determined
by measuring the radius to the tip of the tube of the centrifuge. The
liquid depth will be given by:
Liquid Depth =(Reference RCF at tip - Reference RCF at surface) / Tip * (2)
RCF at Tip ^Radius'
Speeds used in these studies corresponded to RCF values of 259 x g
26
-------
to 313 x g at 1200 rpm, 584 x g to 705 x g at 1800 rpm, 1040 x g to
1262 x g at 2400 rpm, and 1620 x g to 1956 x g at 3000 rpm.
Several centrifugation experiments were run using 8 mg/L of Amoco
cationic polyelectrolyte as an additive. Run time was varied between
one and fifteen minutes, but rotational speed was held constant at
2400 rpm. Run time was calculated from the point that the desired
speed was attained. Thus, acceleration and deceleration time was not
included.
Detartration with Polyelectrolyte Addition -
The pH of California brand stillage is approximately 3.5, and tartrate
is present almost entirely as potassium bitartrate (cream of tartar).
Soluble salts are precipitated as calcium tartrate when calcium
hydroxide added as lime is used to neutralize the stillage in the
presence of soluble calcium ion (as calcium chloride). If calcium
chloride is not present only about one half of the bitartrate will be
removed. The other half will remain in solution as the very soluble
potassium tartrate. The equations below illustrate the necessity for
the addition of soluble calcium salt:
2KHC.H.O, + Ca(OH)0 * CaC.H.O,. + K9C.H.OC + 2H,0 (3)
4 H O C. I Q O £ 4 4 0 c.
K2C4H4°6 * CaC12 * CaC4H4°6 + 2KC1
Tartrate recovery is complicated by the fact that extraneous materials
are also precipitated. This results in contamination of the product.
Where tartrate is to be a saleable product, the process pH should not
be above 4.5 (approximately the iso-electric point of calcium tartrate).
27
-------
If tartrate recovery is not important, precipitation of the extraneous
materials may be advantageous in COD reduction.
Equal weights of calcium chloride and calcium hydroxide (varying from
0.25 grams to 2.25 grams) were added to 500 ml samples of reconstituted
still age. In each case, the mixture was mixed at 100 rpm for 30
minutes and flocculated for five minutes at 20 rpm. Mixing time was
long because the calcium chloride and calcium hydroxide did not readily
dissolve. Filtered COD and pH values were measured.
Polyelectrolyte coagulation of the detartrated still age was also
studied. Purifloc A-23 anionic polyelectrolyte added to the
detartrated stillage (pH 11) in concentrations of 1, 2, 4, 10, 14, 20,
26, 30, 40, 50 and 70 mg/L, mixed at 100 rpm for ten minutes and
flocculated for two minutes at 20 rpm. The mixture was then poured
into a graduated cylinder (500 ml) and interface height and supernatant
liquor COD were measured as functions of time.
Removal of Solids from Supernatant Liquor -
Settled or centrifuged still age still contains high suspended solids
concentrations. Removal of these materials by coagulation with
polyelectrolytes, flocculation and sedimentation, dissolved air
flotation and detartration with polyelectrolyte addition was studied.
The supernatant liquor used was obtained by centrifuging 150 ml
samples of reconstituted stillage for two minutes at 2400 rpm.
Supernatant liquor suspended solids concentration was of the order
of 5800 mg/L. Variables considered in the coagulation of still age
supernatant were cationic polyelectrolyte and bentonite concentrations.
Nalco 610 polyelectrolyte was used in concentrations of 5, 15, 38,
28
-------
60 and TOO mg/L. Bentonite concentration used were 10 and 20 mg/L.
In each case, the coagulants were added to 200 mg/L of still age
supernatant, mixed at 100 rpm for ten minutes and flocculated for
2 minutes at 20 rpm. The mixture was then poured into a 250 ml
graduated cylinder and settling rate and supernatant liquor COD
were measured as functions of time.
Dissolved air flotation was studied using a range of coagulant
concentrations, cell pressures and recycle ratios. The coagulants,
Amoco nonionic and Amoco cationic polyelectrolytes were used in
concentrations of 1, 3 and 7 mg/L. Cell pressures used were 2.1 x 10 ,
2.8 x 106, 3.5 x 106 and 4.1 x 106 dynes/cm2. The recycle fluids
were process supernatant and tap water and ratios of 3.3 and 5 were
used.
Detartration of still age supernatant with polyelectrolyte additive
was studied using the same range of calcium chloride and calcium
hydroxide concentrations used in the still age detartration studies
(500 to 4500 mg/L). In each case, the mixture was mixed for 20
minutes at 20 rpm. Filtered COD and pH values were measured. A
polyelectrolyte (Purifloc A-23) was then added to the detartrated
supernatant liquor (which had a pH value of 11). The mixture was
stirred at 100 rpm for five minutes and flocculated at 20 rpm for
one minute and poured into a 250 ml graduated cylinder. Supernatant
liquor COD and suspended solids were measured after settling.
FIELD SAMPLING PROCEDURES
Samples were taken from the sedimentation influent, activated sludge
mixed liquor and sampling ports of the anaerobic packed beds.
29
-------
s
Settling rate pH and temperature were measured immediately. During
1971 samples to be analyzed for chemical oxygen demand and suspended
solids were packed in ice and immediately transported to Davis for
analysis. During the 1972 operating season suspended solids
measurements were made in the enology laboratory of California State
University, Fresno. Chemical oxygen demand samples were frozen and
analyzed at a more convenient time. This procedure was found to be
acceptable by Chadwick and Schroeder .
Still age used in the laboratory studies was put into clean, 55 gallon
drums, transported to Davis and stored at -55°C until needed. The
laboratory solids removal studies made use of still age which had been
dewatered at the winery, stored at -30°C on the Davis Campus and
reconstituted in small batches by thawing and diluting with five parts
tap water to one part still age concentrate. A reconstituted still age
filtrate chemical oxygen demand value between 18000 and 20,000 was
used as a guide in determining the dilution ratio.
ANALYTICAL METHODS
Temperature was determined with a thermometer in both field and
laboratory studies. Hydrogen ion concentration (pH) was measured in
the field by a portable pH meter, and in the laboratory using a
conventional laboratory instrument (Leeds and Northrup). Dissolved
oxygen measurements were made with a YSI model 5^ portable dissolved
oxygen meter.
Suspended solids were measured using 0.45 micron pore diameter, silver
membrane filters (selas Flotronics - Springhouse, Pennsylvania) and
a conventional membrane filter apparatus. Except for the choice of
30
-------
filters, the method used followed that of the EPA chemical analysis
18
manual .
Chemical oxygen demand was measured according to procedures given in
Standard Methods19.
31
-------
SECTION V
EXPERIMENTAL STUDIES AND RESULTS
1971 PILOT PLANT STUDIES
*
Work on the project between July 30, when the University received
notice that the grant would be awarded, and December 31, 1971 was
entirely related to the Fresno operation. Prior to July 30,
treatment system plans (Figures 3, 4, 5) were drawn up and equipment
could not be made until a project account was set up. Because of
the later than expected starting date unit construction in the shops
was considerably delayed. Two of the five sedimentation-holding
tanks planned and all three of the anaerobic treatment units were
completed early in September and moved to Fresno on September 10th.
The still had been put into operation less than a week prior to the
move and thus the anaerobic treatment processes were in operation
for essentially the entire season. Although completion of the
aerobic units was slow the controlling factor was the arrival of the
gear motors. Five, two horsepower turbines were ordered in August
from a company which stated they were in stock. After repeated
conversations two motors arrived during the last week in September
and the supplier indicated the other three would not be available
until November. The order was then switched to a second supplier,
but the last three gear motors did not arrive until the final week
of October. For this reason only the large aerobic treatment unit
was put into operation.
Because a number of questions needed to be answered before the 1972
operating season a quantity of stillage was stored for use in the
32
-------
laboratory at Davis. Most of still age stored was collected by the
staff of the Gallo Winery and kept under refrigeration until transfer
to Davis was convenient (see Section IV for conditions of storage).
One of the primary reasons a two year project period was proposed
was the assumption that operational nroblems could arise. Because
of the short time period in which the still is in operation any
delays are serious. Four problems were encountered in 1971 at
Fresno: pH control, flow rate control, temperature control, and
removal of suspended solids. The most severe problem was flow rate
control. At no time during the falj was a steady controlled flow
obtained, and therefore the treatment units could not be left
unattended. Whenever the operator left the site for more than a
few minutes he was forced to shut the entire operation down.
Initially the only control device was a valve between the stillage
sewer (a cast iron force main) and sedimentation tank. Flow was
intermittent, even with the valve wide open. Because of the
intermittent flow and the fact that pipe pressure seemed to vary a
pump with a throttling valve was then tried but proved unsatisfactory
also. A major result of the flow problems was that design flow rates
were never reached.
Temperature control was planned for the project and heaters, cooling
coils and control devices were purchased. Unfortunately the heaters
for the anaerobic treatment unit (which was to operate at 50°C) did
not arrive until November, far too late to install them in the units.
Still age temperature at the pilot site varied widely also, and because
of the flow control problem the sedimentation tank was allowed to cool
overnight, every night. Anaerobic process temperatures dropped below
70°F on occasion. Heating coils were used in the sedimentation tanks
33
-------
in November and this improved the situation, raising the temperatures
in the anaerobic units into the low 80's, but not enough to generate
satisfactory fermentation rates.
Control of pH was expected to be a problem, still age acidities
reported in the literature are in the range of 1200 mg/L to 3000 mg/L
(Table 1), and therefore considerable difficulty in neutralization
might be expected. Stillage acidity is due to organic acids which
can be oxidized, and therefore if the mass input rate does not exceed
the oxidation rate pH problems should not result. Because of the low
operating temperatures the anaerobic units were not able to operate
at design rates and pH control, using sodium hydroxide and ammonium
hydroxide (the nitrogen source), was initiated. The program was
successful and the sedimentation tank pH was maintained above 6.0
for most of October and November.
Suspended solids removal is a basic part of the waste treatment
process. Because of the nature of the suspended solids problem with
stillage (high concentration and high fraction with near colloidal
size) biological breakdown was considered a possible method of removal.
Good solids removal is achieved by gravity sedimentation (up to 90%)
if settling velocities of the order of 1.8 to 2.1 meters/day (six to
seven feet per day) are used. Conventional (eg. municipal) treatment
processes utilize a settling velocity range of 32 to 49 meters/day
(105 to 160 ft/day). Sedimentation velocities of stillage are low
because of the high concentration of solids (settling rates are
inversely related to solids concentration), and it should be remembered
that even when 90 percent of the solids have been removed, the
supernatant liquid has a suspended solids concentration on the order
of 1700 mg/L. On a relative volume basis the thickened sludge is
34
-------
about forty to fifty percent of the original volume, thus gravity
settling is not a suitable process.
Biological activity of the treatment processes can be estimated from
the chemical oxygen demand data reported in Tables 3 and 4. Table 3
contains COD data on samples filtered through a 0.45 y filter. The
anaerobic treatment processes were operated as upflow units, thus the
difference between bottom and top filtrate COD readings is equivalent
to filtrate biochemical oxygen demand removed or converted because
the only method of conversion available was biological. As was
stated above, fermentation rates ih the anaerobic processes were
very slow, but fermentation was occurring as shown in Table 3. Tanks
1, 2 and 3 correspond to flow rates of 60, 120 and 240 liters per
hour, respectively.
Organic reductions in the aerobic process were much more satisfactory,
particularly because there was no acclimation of the culture to the
still age. This latter fact is important because of the seasonal
nature of the waste. As was stated earlier, NH^OH was added to the
reactors as a source of nitrogen. The difference between the inlet
tank filtrate COD concentration and the mixed liquor or effluent
filtrate COD concentration (they should be approximately the same)
is a measure of BOD reduction. Previous studies have indicated that
the minimum attainable COD concentration is approximately 1500 mg/L.
Table 4 contains COD data on unfiltered samples. Because process
installation was after the grape harvest began, no attempt to leach
out soluble organics from the wood chips v/as made. COD concentrations
in the anaerobic processes are extremely high because solids accumulate
in the systems. This feature was designed into the processes on the
35
-------
TABLE 3
COD ANALYSIS OF FRESNO SAMPLES (Filtered Samples - 0.45 y) mg/L
Tank 1 Bot.*
Tank 1 Top
Tank 2 Bot.*
Tank 2 Top
Tank 3 Bot.*
Tank 3 Top
A.S. Effluent
Mixed Liquor
Inlet Tank
Raw Waste
10/25/71 10/29/71 10/30/71 10/31/71 11/1/71 11/5/71
17,476
12,122
17,161
14,642
15,114
15,272
14,721
18,505
14,876
17,028
18,674
18,701
16,301
4,985
4,669
15,193
18,620
16,028
17,206
16,145
18,620
16,028
4,478
3,889
15,438
16,263
17,065
16,531
19,392
17,318
21 ,331
15,901
5,825
5,825
14,012
18,657
16,012
15,245
16,683
14,957
15,533
4,698
16,971
16,376
16,473
16,570
16,473
7,073
7,461
15,795
16,473
* Tanks 1, 2 and 3 correspond to inlet flow rates of
60, 120 and 240 liters/hr. respectively.
36
-------
TABLE 3 Cont.
COD ANALYSIS OF FRESNO SAMPLES (Filtered Samples - 0.45 p) mg/L
'Tank 1 Bot.
Tank 1 Top
Tank 2 Bot.
Tank 2 Top
Tank 3 Bot.
Tank 3 Top
A.S. Effluent
Mixed Liquor
Inlet Tank
Raw Waste
11/6/71 11/7/71 11/10/71 11/11/71 11/15/71
16,695
14,419
15,936
14,703
16,695
6,166
6,166
13,755
15,557
14,824
16,823
15,813
17,631
16,722
6,220
6,523
13,995
15,409
18,093
17,193
17,193
16,693
17,893
16,993
5,598
6,097
16,194
15,594
17,066
16,056
16,460
15,854
17,873
17,167
6,665
6,867
17,066
14,844
18,601
15,435
16,721
17,018
19,491
15,534
4,452
4,452
13,159
37
-------
TABLE 4
COD ANALYSIS OF FRESNO SAMPLES (Unfiltered Samples) mg/L
Tank 1 Bot.*
Tank 1 Top
Tank 2 Bot.*
Tank 2 Top
Tank 3 Bot.*
Tank 3 Top
A.S. Effluent
Mixed Liquor
Inlet Tank
Raw Waste
10/25/71 10/29/71 10/30/71 10/31/71 11/1/71 11/5/71
29,700
36,748
15,360
33,377
48,274
34,164
12,359
14,170
34,401
51,198
35,339
58,299
35,458
14,864
15,575
31,197
33,351
56,043
35,669
74,768
65,272
34,472
13,875
16,776
38,334
67,407
41 ,773
83,502
44,417
62,171
40,304
17,568
18,703
30,904
30,806
48,723
34,191
45,685
33,711
14,439
16,932
35,150
28,534
* Tanks 1, 2 and 3 correspond to flow rates of 60, 120
and 240 liters/hr.
38
-------
TABLE 4 Cont.
COD ANALYSIS OF FRESNO SAMPLES (Unfiltered Samples) mg/L
Tank 1 Bot.
Tank 1 Top
Tank 2 Bot.
Tank 2 Top
Tank 3 Bot.
Tank 3 Top
A.S. Effluent!
Mixed Liquor
Inlet Tank
Raw Waste
11/6/71 11/7/71 11/10/71 11/11/71 11/15/71
60,947
32,617
64,594
38,721
17,985
17,500
34,458
17,500
82,598
24,737
77,091
34,613
68,737
28,285
14,670
14,574
16,499
16,683
77,712
36,413
73,312
41 ,462
60,714
33,081
16,621
18,136
30,759
32,980
65,113
42,674
64,114
44,390
71,112
49,036
15,712
20,660
28,941
90,666
23,091
63,545
47,980
63,545
30,788
15,794
13,395
12,595
39
-------
assumption that solids degradation would occur during the off season.
If a significant degradation does occur a solution to the solids
degradation problem using nominal amounts of land will be available.
Data presented in Table 4 for the aerobic process is of interest
because the difference between inlet tank COD and effluent COD is
greater than that for the filtrate. Thus some breakdown and
oxidation of suspended solids occurred. Inlet tank COD can be seen
to be about fifty percent in the filtrate and fifty percent in the
suspended solids. The data in Tables 3 and 4 indicates that
approximately one third of the suspended solids were removed.
1972 LABORATORY STUDIES
Aerobic and anaerobic biological treatment studies were run on a
laboratory scale during the winter of 1972. Experimental systems
and procedures were described in Section IV. Settled still age
supernatant liquor was used as a feed for both aerobic and anaerobic
systems. Still age settling rate varies from sample to sample, but
values are uniformly low. Results of a typical settling test on raw
still age are shown in Figure 6. Settling rates measured in the field
2 2
varied from about 0.18 L/cm -day (45 gal/ft -day) down to zero (no
clear-water-solids interface formed and very little solids
accumulation on the cylinder bottom after a 24 hour period).
Supernatant liquor used in the laboratory studies was obtained by
filling 20 liter carboys with thawed still age and allowing the
material to settle for 24 hours.
Supernatant liquor COD and suspended solids concentrations were
approximately 19,000 mg/L and 12000 mg/L, respectively. Influent
40
-------
£2
LU
o
U-
tr,
280
224 -
168 -
112 -
56
TIME , HOURS
FIGURE 6. TYPICAL RAW STILLAGE SETTLING CURVE
41
-------
E
2
UJ
CD
X
O
O
20
18
16
14
o 12
10
I
_L
FILTERED
I
I
1234
HYDRAULIC DENTENTION TIME, (DAYS)
FIGURE 7. STILLAGE FEED CONCENTRATION vs. HYDRAULIC
DENTENTION TIME FOR LABORATORY REACTOR
42
-------
characteristics were constant for each experiment (i.e. each residence
time, but varied somewhat throughout the study).
Aerobic Treatment Studies
Aerobic process influent COD concentrations (to the first reactor in
the series) are shown in Figure 7 as a function of residence time.
The processes were monitored at a given detention time until steady
state conditions were attained. This normally took between one and
two weeks. Steady state data reported here was taken over a fourteen
day period in each case. Under the conditions imposed on the
treatment units pH was an excellent indicator of process performance.
The low pH and high acidity of stillage is due, primarily, to organic
acids. As long as the organic acid conversion rate equals the input
rate pH remains near neutral. When input rate exceeds the conversion
rate effluent COD concentrations increase and pH values decrease
because of the uncoverted organic acids present.
Total reactor suspended solids concentrations are given in Table 5.
Effluent quality, as measured by COD concentration was not improved
by using the series reactor system. Operation was stable in the
first reactor system at residence times down to 1.4 days, although sludge
settling deteriorated below values of 1.87 days. Settled effluent
COD values were virtually the same for both units. Dissolved oxygen
concentrations reflected the relative activity of the two units, as
was mentioned previously pH was not controlled, but instead was used
as a measure of overloading. Values of pH, dissolved oxygen and
effluent COD concentrations are reported for the first reactor in
Table 6 and Figure 8, and for the second reactor in Table 7 and
Figure 9. Biochemical oxygen demand (BOD5) of the filtered effluent
43
-------
was approximately 75 mg/L.
Removal rates, cell growth rates and cell yield are important
parameters in process design. Normally the rates are believed to
be linear functions of the cell mass concentration and this allows
the use of specific or unit rates (rate per unit mass of cells).
Because the suspended solids concentration of the settled stillage
v/as high (^1200 mg/L) the mixed liquor suspended solids concentration
(MLSS) could not be assumed to represent the cell mass concentration,
and thus could not be directly used in calculating unit rates. An '
estimate of the cell mass concentration was made by subtracting the
stillage suspended solids concentration from the MLSS concentration.
Stillage solids include grape pulp, bits of stems and leaves and
yeast cell residues, all of which have been broken up during the
distilling process. Thus the material remaining in nonsoluble form
can be assumed difficult to degrade. Subtraction of the settled
still age SS from the MLSS concentration and using this estimated cell
mass concentration to calculate the unit removal rate results in the
information presented in Figure 8. Unit growth rate in a well mixed
unit without cell recycle is equal to the inverse of the hydraulic
residence time. The ratio of the two rates (growth rate/removal
rate) is the maximum cell yield, and as noted on Figure 10, the
value is 0.37 grams cells produced per gram of COD removed.
Het cell yield is not a constant due to increasing cell maintenance
energy requirements with increasing cell age. Estimation of the
maintenance energy requirement is difficult with the data presented
here because of the high growth rates used in the study and the
method of determination of cell mass concentration. Extension of the
44
-------
TABLE 5
MIXED LIQUOR SUSPENDED SOLIDS CONCENTRATIONS, mg/L
8; days Reactor 1 Reactor 2
1.09 5600 7400
1.41 8400 6800
1.87 8700 6700
2.42 9200 6900
2.75 8700 6900
3.58 7100 5900
4.67 7200 6100
45
-------
TABLE 6
OPERATIONAL VARIABLES FOR FIRST LABORATORY REACTOR IN SERIES
, days
1.09
1.41
1.87
2.42
2.78
3.58
4.67
OPERATIONAL
8, days
1.09
1.41
1.87
2.42
2.75
3.58
4.67
DO
mg/L
1.9
3.1
3.2
No
Data
3.7
5.2
6.6
VARIABLES
DO
mg/L
1.9
6.7
6.4
-
7.6
7.7
7.7
PH
5.1
6.5
6.3
6.3
6.1
6.2
6.6
FOR
pH
7.3
6.4
6.2
6.1
5.9
5.7
6.1
Effluent
Settled
12,079
2,264
1,484
3,303
2,380
1,484
—
TABLE 7
SECOND LABORATORY
Effluent
Settled
3,149
1,287
1,552
2,839
2,149
1,450
No Settl
COD, mg/L
Filtered
7,009
636
592
763
643
537
666
REACTOR /w SERIES
COD, mg/L
Filtered
1,031
664
773
663
608
554
ing 724
46
-------
&
a"
ae
12
10
X c
o 6
o
SETTLED
FILTERED
o -o
I
012345
HYDRAULIC DETENTION TIME, (DAYS)
FIGURE 8. EFFLUENT COD CONC. vs. HYDRAULIC DETENTION
TIME FOR LABORATORY REACTOR 1
47
-------
4000
- 3000
O
O
O
2000
1000
I
I T
O SETTLED
FILTERED
I
1
1
012345
HYDRAULIC DETENTION TIME, DAYS
FIGURE 9. EFFLUENT COD CONC. vs. HYDRAULIC DETENTION
TIME FOR REACTOR 2 IN SERIES
48
-------
curve in Figure 10 to the y axis gives a value of 0.01 day" for
the maintenance energy coefficient, k. , in the equation below:
- (Y* rox
1
0
where y is the unit growth rate and r is the unit removal rate.
OX
Note that r is inherently negative.
Anaerobic Treatment Studies
Methods and procedures used in the laboratory anaerobic treatment
studies were described in Section IV. As was noted the feed COD
concentration to these systems was 15,500 mg/L.
Results of the anaerobic experiments were not promising. The 15 day
residence time unit failed completely, producing very little gas and
having an effluent COD of about 14,000 mg/L. At a 30 day residence
time COD concentrations were reduced to an average value over a two
month period of 5100 mg/L with a gas production of 500 ml/day and a
pH of 6.9. The gas production corresponds to 15 ft. of gas per pound
of COD removed. Operation at a residence time of 60 days did not
improve COD conversion measurably, and of course gas production rate
decreased proportionately (to approximately 250 ml/day).
1972 PILOT PLANT STUDIES
Unfortunately 1972 was a poor year for grapes in the San Joaquin
Valley. The harvest season was very short and stillage production
extended only from the last week of August to the second week of
49
-------
en
o
CQ
o
oc
CD
o
o
LLJ
Q_
1.0
0.8
0,6
0.4
0.2
T — r — i — i — i — i — i — i
i — r
YIELD COEFFICIENT, Y=0.37
i i
I I I I I I I
0 0.2 0,6 1.0 1.4 1-8 2.2
UNIT SUBSTRATE REMOVAL RATE, (days'1)
2,6
FIGURE 10. SPECIFIC GROWTH RATE VERSUS UNIT
SUBSTRATE REMOVAL RATE
-------
October. The pilot plant systems were set up as shown in Figure 5
but operational problems with the small aerobic units (primarily
electrical problems with the motors) prevented extensive operation
of the entire system.
Loading rates used during the 1972 field studies were chosen on the
basis of the laboratory results. Aerobic processes were operated
without recycle at a residence time of 3 days. Because the 1971
anaerobic studies run at very short hydraulic residence times were
unsuccessful the 1972 studies were run at 3 day residence times.
Nitrogen was not added to the still age in 1972. Stoichiometric
quantities needed would be of the order of 1000 mg/L. Assuming
nitrogen present in the still age is available for synthesis
approximately 250 mg/L would still have to be added. This
corresponds to 3230 mg/L of ammonium chloride or 2130 mg/L of
ammonium nitrate per liter of still age. For a winery as large as
Gallo-Fresno, this would mean adding over 4000 kilograms (9000/lbs.)
per day. Because of the large nitrogen requirement it was decided
to determine the extent of treatment possible without nitrogen
addition. This resulted in a COD/N ratio of approximately 60:1.
A number of operating problems associated with the characteristics
of the still age also occurred. These problems together with the
extremely short 1972 operating season hampered data collection.
Two problems of significance were difficulties in pumping still age
and foaming resulting from agitation in the aerobic processes.
Pumping difficulties resulted from the impeller sizes used. Clogging
by stems and debris was difficult to control, but would not be a
51
-------
problem with different pumps. Foaming was a far more significant
problem and resulted in considerable problems, both for the pilot
plant operation and for the winery. Foam layers built up to the
point that overflow of the aeration tanks occurred on occasion.
In addition the attached secondary clarifier units (which were not
used, but through which the flow passed by using a bottom drawoff)
developed a layer of thick stable foam which often overflowed the
tank also. The foam restricted oxygen transfer by the surface
aerators with resulting odor problems, and proved to be an ideal
breeding area for flys. Screens were placed over all of the units
except the aeration tanks and insecticides were applied. Gallo
provided advice and help on the problem, but control was never
completely satisfactory. Because of the proximity of the experimental
area to the winery there was considerable concern that the fly problem
would cause action by the county health officer.
The anaerobic treatment units did not function well during the 1972
season but were improved over 1971. Installation of heaters into
i
the anaerobic processes was impossible without complete draining and
media removal. Because of the difficulty of this process, it was
decided to place the heaters in the sedimentation tanks. Lack of
temperature control was again the primary factor in the poor results,
and thus data reported here is qualitative in nature. During the
1971 season which extended from September 5th to November 15th
virtually no anaerobic treatment took place as has been noted although
the units proved to be excellent sedimentation tanks. During the
1972 season, COD removals were greatly improved. Effluent filtrate
COD values as low as 3900 mg/L and ranging up to 7500 mg/L were
obtained with influent COD values ranging from 16,000 to 19,000 mg/L.
Control of pH was maintained in the pilot units by cutting off
52
-------
Influent flow whenever pH values dropped below 6.5.
Startup of the anaerobic units during the 1972 season was done
without adding new cells. The units were left unattended from
November 15, 1971 until early August, 1972 when water was added to
make up for evaporation. Because of the large quantities of organic
solids in the tanks at the end of the 1971 season it was felt that
an improved culture would develop during the nonoperational period.
Evidently a culture did develop because there were no start up
problems and COD conversion was much improved as noted previously.
Aerobic treatment results must be evaluated in light of the lack of
available nitrogen. Effluent filtrate COD values ranged from 1460
to 7320 mg/L with an average value of 4600 mg/L. Settled effluent
COD values were about 1800 mg/L greater than filtrate COD values.
Suspended solids concentrations in the aeration tank varied with the
settleability of the incoming still age. Values ranged from 7500 mg/L
to 14,000 mg/L. Data for 1972 correlated very well with that obtained
in 1971.
Significant start up problems did not occur either year, and pH
control was not a problem as long as the system was not organically
overloaded. The major operational problems were still age solids
during periods when still age settleability was poor, and foaming.
Both solids and foaming retard oxygen transfer and thus decrease
process efficiency and can result in anaerobic conditions.
POMACE FERMENTATION
Pomace fermentation is conventionally carried out after mixing with
53
-------
stems in the disintegrator and washing with water. The pomace is
partially fermented prior to this operation, but residual sugars and
sugars washed off of the stems are converted to alcohol at this
point. Alcohol concentration is low (1 to^4%) but economic recovery
is possible. Because most of the potential alcohol is in the pomace
the possibility exists for fermenting the pomace only (leaving out
the stems and washwater), and greatly decreasing the wastewater flow
rate.
Temperature control is the primary concern because the material
being fermented would be the pressed pomace (Figure 1) which is
similar to a filter cake. The purpose of the experiments in this
study was to determine the amount of heat generated during
fermentation and propose a method of dealing with the problem.
Temperature rise can be controlled by controlled dilution with water
providing mixing and cooling (a tumbler type device) or by blowing
inert gases through the pomace (eg. water saturated N9 or C0«).
L* Cm
Because of the operational problems encountered at Fresno during the
1972 season there was not time enough to run the experiments at the
pilot plant site. Material was stored and experiments were then
conducted at Davis after the processing season. Unfortunately these
experiments failed and there was no way to repeat them. Useable data
was not obtained from these experiments.
COAGULATION STUDIES
Coagulants used were limited to polyelectrolytes because of the low
pH and high acidity of the stillage. Although a generally negative
charge on the particles was believed to exist, anionic and nonionic
polymers were investigated. These included Purifloc A-23, Amoco
54
-------
Anionic, Separan MG 200, Separan MGL, Purifloc N-12, Nalco 607,
Nalco 634, Nalco 671, Nalco 676, Bentonite, and Amoco Nonionic
Coagulation was not induced in any of the experiments with the
polyelectrolytes.
Cationic polymers used included Purifloc C-31, Purifloc C-41, Amoco
Cationic, Nalco 610 and Nalco 610-HD. Jar tests with still age were
run using concentrations ranging from 10 to 200 mg/L of each
polyelectrolyte. Coagulation occurred with all of the materials,
but the best results were obtained vdth Purifloc C-41 and Amoco
Cationic. Settling tests were then run on a 1/2, 2/3 by weight
mixture of Purifloc C-41 and Amoco Cationic at concentration values
of 7, 15, 24, 36 and 60 mg/L. Results are shown in Figures 11 and
12. Data points were omitted in Figure 11 because of the number of
curves and the fact that scatter was nonexistant (i.e. curves were
drawn from point to point).
CENTRIFUGATION
Results of the raw still age centrifugation studies are shown in
Figures 13, 14 and 15. Satisfactory sludge concentration was
obtained even at the lowest speed (see Figure 15) where the cake
moisture content and suspended solids removal were both approximately
88% after five minutes. Cake moisture content decreases with time
because the cake becomes more compressed. Thus although removal of
suspended solids increases with time, the sludge volume does not
necessarily increase. For example, when suspended solids removal
goes from 86% at 2 minutes to 93% at 15 minutes for the 120 rpm
curve, the cake moisture decreases from 88% to 87.1% and the
corresponding sludge volume (based on an initial suspended solids
55
-------
o
u.
or
0 400 800 1200 1600
TIME , MINUTES
FIGURE 11. SETTLING CURVES FOR PURIFLOC
C-41 AMOCO CATIONIC MIXTURE,
56
-------
- 300
o
C/3
a
LU
o
Q_
t/9
200
100
T
T
T
RAW STILLAGE COD: 42,400 mg/1
FILTERED STILLAGE CID: 17,900 mg/l
SETTLED STILLAGE COD: 19,500 mg/l
SUSPENDED SOLIDS OF SETTLED STILLAGE
1270 mg/l
• SUSPENDED SOLIDS
O SUSPERNATANT LIQUOR COD
I I L
1
10
20 30 40
COAGULANT ADDED, mg/l
50
20
18
LO
o
o
16
14
60
FIGURE 12. EFFECT OF COAGULANT ADDITION ON SUPERNATANT LIQUOR
-------
95
E 90
o
o
85
80
RAW STILLAGE COD: 36,700 mg/l
1
1
1200 RPM
1800 RPM
2400 RPM
3000 RPM
I
0 5 10 15
CENTRIFUGE RUN TIME, MIN
FIGURE 13. CAKE MOISTURE vs. CENTRIFUGE
RUN TIME
58
-------
100
95
o
cc
S 90
O
LU
O
85
80
3000 RPM
2400 RPM
1800 RPM
1200 RPM
RAW STILLAGE SUSPENDED
SOLIDS CONC.: 12,000 mg/l
1
1
I
05 10 15
CENTRIFUGE RUN TIME , MIN.
FIGURE 14. SUSPENDED SOLIDS REMOVAL VERSUS
CENTRIFUGE RUN TIME
59
-------
S, 20
o
o
S 19
oc
o
O1
18
QC
UJ
Q_
CO
17
RAW STILLAGE COD: 36,700 mg/l
I
3000 RPM
I
1
0 5 10 15
CENTRIFUGE RUN TIME , MIN.
FIGURE 15. SUPERNATANT LIQUOR COD VERSUS
CENTRIFUGE RUN TIME
60
-------
concentration of 11,952 mg/L) increases from 75.2 to 75.7 mg per
liter of waste. At 3000 rpm, the sludge volume after one minute run
time is 66.4 ml/L of waste. This decreases to 54.9 ml/L of waste
after 15 minutes.
Chemical oxygen demand values also decreased sharply after centri-
fugation. Rotational speed was held constant at 2400 rpm
(corresponding to a RCF value of 1040 x g to 1262 x g). Run times
used were 1, 3, 6, 10 and 15 minutes. Amoco cationic polyelectrolyte
v/as used at one concentration, 8 mg/L. Cake moisture (in %), COD,
and suspended solids concentration were measured as functions of time.
Results are shown in Figures 16, 17, 18.
Results of the experiments on detartration of the raw stillage
followed by polyelectrolyte addition are shown in Figures 19 and 20.
Figure 19 shows the relationship between the quantity of calcium
hydroxide and calcium chloride added and the resulting supernatant
liquor pH and COD. The curves in Figure 20 are the result of adding
an anionic polyelectrolyte (Purifloc A-23) to the detartrated
stillage at a pH of 11. Polyelectrolyte was added in concentrations
of 1, 2, 4, 10, 14, 20, 26, 30, 40, 50 and 70 mg/L, but only four
curves are shown in Figure 20 because many of the curves fell on top
of one another. It should be noted also that the ordinate in
Figure 20 does not go to zero.
REMOVAL OF SOLIDS FROM SUPERNATANT LIQUOR
Solids concentration and removal methods for pomace stillage
supernatant were: a) coagulation with polyelectrolytes, flocculation
and sedimentation, b) dissolved air flotation with and without
61
-------
86
I 84
UJ
5 82
80
T
2400 RPM
8 mg/l AMOCO CATIONIC
I
I
0 5 10 15
CENTRIFUGE RUN TIME, MIN.
FIGURE 16. CAKE MOISTURE vs. CENTRIFUGE
RUN TIME
62
-------
98
98
S 96
o
UJ
DC
Q
Ij
O
Q
UJ
LU
a.
94
92
2400 RPM
RAW STILLAGE SUSPENDED
SOLIDS CONC.: 23,400 mg/l
I
1
I
5 10 15
CENTRIFUGE RUN TIME, WIN.
FIGURE 17. SUSPENDED SOLIDS REMOVAL VERSUS
CENTRIFUGE RUN TIME
63
-------
22
KJ
eb
cT
O
O
oz
O
21 21
oe
LU
a.
20
1
I
I
5 10 15
CENTRIFUGE RUN TIME, MIN.
FIGURE 18. SUPERNATANT LIQUOR COD VERSUS
CENTRIFUGE RUN TIME
64
-------
• pH
o FILTRATE COD
- 14
- 13
- 12
- 11
10
o
o
1.0 2.0 3.0 4.0
CaCI2, Ca(OH)2Conc., g/l
FIGURE 19. SUPERNATANT LIQUOR pH AND FILTRATE
COD vs. CaCI2 and
65
-------
300
250
O
LU
LLJ
O
200
150
100
50
I I
COAGULANT: PURIFLOCA-23
30mg/l
50mg/l
\ "1
\
\
\
\
I
100
150
200
hrs
TIME , MINUTES
FIGURE 20. SETTLING OF DETARTRATED RAW STILLAGE
66
-------
polyelectrolyte addition, and c) detartration with polyelectrolyte
addition. The supernatant was obtained by running 150 ml samples
of reconstituted raw still age through the IEC centrifuge for two
minutes at 2400 rpm. Supernatant liquor suspended solids concentration
was of the order of 5800 mg/L.
Coagulation
Variables considered in the coagulation of the stillage supernatant
included cationic polyelectrolyte and bentonite concentrations.
Nalco 610 polyelectrolyte was used at concentrations of 5, 15, 35,
60, and 100 mg/L. Bentonite concentrations were 10 and 20 mg/L.
In each case the coagulants were added to 200 ml of still age
supernatant, mixed at 100 rpm for ten minutes and flocculated for
two minutes at 20 rpm. The mixture was then poured into a 250 ml
graduated cylinder. Settling properties and supernatant COD were
measured as function of time. Settling times were less than five
minutes in all cases, and an interface did not form (i.e. settling
was not hindered). Results are shown in Figure 21.
Dissolved Air Flotation
Variables considered in the dissolved air flotation studies included
coagulant, air concentrations, recycle fluid, pressure and recycle
ratios. Cell pressures and recycle ratios used were 2.07 x 10 ,
2.76 x 106, 3.45 x 106, 4.14 x 106 dynes/cm2 (30, 40, 50, and 60 psig)
and 3.33 and 5.0, respectively. Water and recycled supernatant were
used as the recycle fluid. Amoco nonionic and Amoco cationic
polyelectrolytes were used in 1, 3, and 7 mg/L concentrations. Runs
67
-------
20
O
O
O
O
O
19
18
17
16
15
I I I
RAW STILLAGE COD: 39,000 mg/l
• 10 mg/l BENTONITE
020 mg/l BENTONITE
^ FILTERED RAW STILLAGE
25 50 75
NALCO 610 CONC,. , mg/l
100
FIGURE 21. SUPERNATANT COD vs. COAGULANT CONC,
68
-------
on the still age supernatant without aids were also run. Satisfactory
solids separation did not occur in any of the experiments.
Detartration With Polyelectrolyte Addition
Tartrate in still age is in a soluble form as was noted in Section IV.
Thus, detartration of the stillage supernatant liquor rather than
raw still age has the advantage of less contamination of the
precipitate. Quantities of calcium chloride and calcium hydroxide
necessary would be expected to be similar, as is shown in Figure 22.
Addition of an anionic polyelectrolyte (Purifloc A-23) improved both
suspended solids and COD removal as shown in Figure 23.
69
-------
14
12
10 —
0 PH
o COD
RAW STILLAGE COD: 38,000 mg/l
8 —
6 -
1.0 2.0 3.0
CaCi2,Ca(OH)2 , mg/l
2
ab
O
O
O
4.0
FIGURE 22 pH AND COD VERSUS CaCI2 and Ca(OH)2 CONC.
70
-------
^ 4000
o
o
O
tst
a
UJ
a.
no
3000
2000
1000
cnnn i • SUSPENDED SOLIDS
5000 *- O COD
RAW STILLAGE COD: 28,700 mg/l
I
I
1
25 50 75
PURIFLOC A-23 CONC., mg/l
15
14
13
12
o
o
o
11
10
FIGURE 23 . SUSPENDED SOLIDS AND COD CONC. OF
DETARTRATED SETTLED SUPERNATANT LIQUOR
71
-------
SECTION VI
DISCUSSIO
-..ri
Biological treatment of California winery stillage is possible,
either anaerobica.lly or aerobically. In either case additional
treatment will be necessary. Aerobic treatment will require that a
significant fraction (of the order of 902) of the suspended solids
be removed prior to aeration. Suspended solids removal may not be
necessary prior to anaerobic treatment, but the high effluent
organic concentrations (COD concentrations of the order of 5000 mg/L)
associated with anaerobic treatment will force the use of an
additional, probably aerobic, treatment process.
Because of the seasonal nature of the waste aerobic processes seem
more desirable than anaerobic processes. Aerobic process "start up"
time is short and the results of the pilot and laboratory studies
lead to the conclusion that a successful treatment system can be
designed and constructed. Aerobic treatment of settled stillage can
produce effluents with filtrate COD and BODr values of the order of
700 mg/L and 75 mg/L, respectively. Settling rates of the activated
sludge are very low, even after further aeration in secondary units,
and the quantity of nonsettleable material is relatively high.
Thus5 while aerobic treatment does an excellent job of converting
organic material, the residual effluents COD and suspended solids
concentrations are still unsatisfactory. Finally, the quantities of
nitrogen which must be added are extremely high. Ammonium chloride
should not be used because of the high quantity of chlorides which
would be added to the effluent. Less ammonium nitrate is needed on
a pound per gallon basis, but ammonium nitrate is about twice as
72
-------
expensive as ammonium chloride. Phosphorous is available in excess
and will not need to be added.
Successful aerobic biological treatment will be dependent upon
pretreatment for solids removal. Successful solids removal will
involve concentration from about two percent to ten percent. This
is necessary to minimize the solids volume to reasonable proportions.
Several methods of treating- the solids are possible including
anaerobic digestion. Selection of a method of suspended solids
disposal was beyond the scope of this project.
Based upon the laboratory and pilot plant studies results, a system
of aerated lagoons is recommended as the best method of biological
treatment of winery stillage.
£'-. v
Following solids removal, aerated lagoons designed for a three to
three and a half day residence time should be used to remove most
of the organic material. In order to reduce the nitrogen requirement
nitrogen should not be added to these aerated lagoons, but instead
••; •-
should be added to;the effluent as it flows into a second set of
aerated lagoons. Based on the 1972 pilot studies the organic
concentration will be about 20 to 25 percent of that in the settled
raw still age, allowing a correspondingly lower addition of nitrogen.
Effluent from the second set of aerated lagoons should be allowed
to settle in holding ponds with a minimum of one day residence time.
Settled solids can collect on the bottom and degrade during the
nonoperating months as in conventional stabilization ponds. A
schematic of the proposed process is shown in Figure 24. Effluent
from the final ponds should be suitable for irrigation or possibly
for discharge into municiple sev/ers.
73
-------
SOLIDS
REMOVAL ,
y^i
-------
Coagulation, flocculation and sedimentation of still age appear to be
an unsatisfactory solids removal method because of the sludge volume
produced. In all of the experiments sludge volume exceeded one third
of the original liquid volume and therefore use of this process would
simply create two disposal problems from the initial one. The fact
that raw still age had better settling properties than the coagulated
stillage is probably due to the increase in particle interaction with
flocculation. A point should be made that supernatant suspended
solids concentrations are much lower in the coagulated stillage than
in the untreated still age (Figure 12). Chemical oxygen demand values
are also less but this is probably primarily a result of the improved
solids removal.
Centrifugation proved to be the best method of removing solids from
the still age. Sludge volume was satisfactory (approximately 10%)
and removals can be achieved at feasible speeds and run times.
Operating a continuous flow centrifuge under conditions to match the
batch data at five minutes and 1800 rpm would produce a product of
approximately 1000 mg/L suspended solids and 18,000 mg/L COD.
Reardon4 found that still age of this strength was treatable in
aerated lagoons. Cake moisture under these conditions is approximately
8%, and the corresponding sludge volume would be 9% of the original
volume. It should be noted that cake moisture content is an average
value and thus tends to decrease as lighter, less compactable solids
settle out at longer run times.
The effect of run time on cake moisture content is even more
noticeable when coagulants are used. Improved suspended solids
removal occurs, increasing the total amount of solids in the cake and
the cake volume. Cake moisture content decreases more sharply than
75
-------
with the untreated still age in this case, as would be expected.
Detartration is of interest if potassium bitartrate is to be recovered.
Results of the detartration experiments on raw stillage lead to the
conclusion that the process is not suitable. Three problems are
associated with this procedure, final separation of the tartrate,
^
the very low settling rates which develop and the large sludge volume
that would result, even when high polyelectrolyte concentrations are
used.
Studies on solids removal from supernatant liquor were undertaken to
determine what quality of effluent can be achieved with respect to
solids removal. Pretreatment by centrifugation was chosen because
this appeared to be the most effective treatment process for the raw
stillage. Coagulation with a polyelectrolyte and bentonite worked
extremely well on this less concentrated material and the batch
settling time was less than five minutes in all cases.
Detartration of the supernatant liquor is much more straight forward
than in the case of the raw still age, particularly with respect to
solids removal. As in the case of coagulation of the supernatant
liquor settling rates were high (batch times of the order of two
minutes or less and no interface was formed).
*'•
Foaming problems will be less than those experienced in the pilot
studies in larger ponds. Good solids removal will also remove much
of the light pulpy material which gave the pilot study foam the
properties most difficult to deal with.
Biological treatment of winery stillage is an alternative to intermittant
76
-------
irrigation. Prior to any changes in the present method of disposal,
intermittent irrigation should be studied further. The possible use
of tile drainage to collect the wastewater should be considered,
together with study of the quantity and quality of water actually
moving through the soil. If nuisance problems exist some effort
should be made to develop a systematic method of application and of
nuisance control.
Additional study of anaerobic treatment is not recommended. The major
advantages of anaerobic treatment are the utilization of the high
temperature effluent from the still, the small amount of nitrogen
addition necessary (not established in these studies but probably of
the order of 10% of that stoichiometrically needed for aerobic
processes) and the methane gas production. Major disadvantages include
the long start up times, and the requirement of additional treatment
of the effluent. The start up problem is extremely important and the
fact that the aerated lagoon systems proposed appear satisfactory
leads the conclusion that aerobic processes are preferable.
ESTIMATED CAPITAL COST
;
The proposed treatment process would include solids removal by centri-
fugation, biological treatment without nitrogen addition in an aerated
pond with a three day residence time followed by biological treatment
in an aerated pond with nitrogen addition and a three day residence
time and finally a clarification pond to remove and store solids, as
shown in Figure 24.
The centrifugation step will be a major cost item. A continuous cake
discharge, solid bowl, scroll type centrifuge ranges in cost from
77
-------
approximately $25,000 for a machine capable of handling 55 liters/min.
(14.5 gpm) to approximately $100,000 for a unit capable of handling
475 liters/min. (125 gpm). A very large winery (eg. Gallo-Fresno)
therefore would need about four large centrifuges with a capital outlay
of approximately $400,000.
fi o
Basin cost for a 2.3 x 10 liter/day (6 x 10 gpd) plant assuming
four identical aeration basins, each six feet deep, and having a total
volume of 13,655 cubic meters (17,800 cubic yards) and two settling
basins, also six feet deep and having a total volume of 2300 cubic
meters (3000 cubic yards) is approximately $57,000 or $25/1000 liters/
day ($24/3785 gpd). Costs are based on $2.10 per cubic meter for
excavation and $2.63 per cubic meter for basin walls.
Floating aerators would be the best choice for the system described.
Considering the expected cell yield and the average settled still age
COD values the expected oxygen demand is approximately 1136 kilograms
per hour (2500 Ib/hr) for a 2.3 x 10" liter/day flow rate. Aerators
up to 112 kilowatts (150 hp) in size are available and considering the
high power to volume which will be achieved oxygen transfer rates of
1.22 gram/watt-hr (2 Ib/hr-hr) can be expected. Approximately nine
aerators at an installed cost of about $30,000 each would be required.
Thus the total cost of centrifuges, aerators and basins for a 2,300,000
liter per day treatment plant would be approximately $730,000.
Piping, pumps and laboratory space and equipment must be added to this
total. Operation and maintenance are not included either. Figure 25
presents estimated costs for the three major capital items as a
function of flow rate. Consideration of the total cost of treatment,
78
-------
o
o
70
60
50
40
30
20
10
I I I I
d
0 0.5 1.0 1.5 2.0 2.5
FLOW RATE, Liters/Day x 10~6
FIGURE 25. ESTIMATED TOTAL COST OF CENTRIFUGES
AERATORS AND BASINS.
79
-------
the seasonal nature of winery operations and fact that treated effluents
will probably be used for irrgation wherever possible leads to the
conclusion that biological treatment is a less satisfactory process
than direct land disposal by internrittant irrigation. The primary
concern with respect to intermittent irrigation is the possible
contamination of the soil and groundwater with salts. Biological
treatment of still age is feasible and will be suitable in cases where
suitable land is not available for irrigation. Properly designed
processes should produce a product suitable for use as irrigation water
or for discharge into a municipal sewer.
80
-------
SECTION VII
REFERENCES
1. Coast Laboratories. The disposal of winery wastes. Progress
Report to the Wine Institute, San Francisco. (1946)
2. Pearson, E. A., D. F. Feuerstein, and D. Onodera. Treatment
and utilization of winery wastes. Proceedings of the 10th
Industrial Waste Conference, Purdue University. (1955)
3. Chadwick, T. H., and E. D. Schroeder. Characterization and
treatability of pomace still age. Journal of the Water Pollution
Control Federation, 45, 1978, (1973).
4. Reardon, D. J. Aerobic Treatment of Pomace Still age. M.S.
Thesis, University of California, Davis, Department of Civil
Engineering. (1972)
5. Stander, G. J. Treatment of wine distillery waste by anaerobic
digestion. Proceedings from the 22nd Industrial Waste Conference,
Purdue University. (1967)
6. South African National Institute for Water Research. Investigation
of the full scale purification of wine distillery wastes by the
anaerobic digestion process. C.S.I.R. Research Report No. 270,
UDC-628, Cape Regional Laboratory, Bellvilie, South Africa. (1968)
7. Jeperson, Paul. Personal communication. California Regional Water
Quality Control Board, Central Valley Region. (September, 1972)
8. Amerine, M. A., H. W. Bert and W. V. Creuss. The Technology of
Wine Making, 2nd Edition. The AVI Publishing Company, Inc.,
Westport, Connecticut. (1967)
9. Schmidt, K. D. The Distribution of Nitrate in Ground-Water in
The Fresno-Clovis Metropolitan Area, San Joaquin Valley,
California. Dissertation for the Degree of Doctor of Philosophy,
University of Arizona. (1971)
81
-------
10. York, 6. K. Land disposal of stillage - analysis of soil core
samples. Report to Wine Institute. (November 8, 1972)
11. Vaughn, R. H., M. S. Nightingale, J. A. Pridmore, E. M. Brown,
and G. L. Marsh. Disposal of wastes from brandy stills by
biological treatment. Wine and Vines, 31(2):24. (1950)
12. Hodgson, H. 6., and J. L. Johnston. Disposal of wastes at
Glenely, South Australia. Sewage Works Journal, 12:321. (1940)
13. Ingram, W. T. Treatment of winery wastes by aeration and other
methods. Report to the Wine Technical Advisory Committee.
(May, 1961)
14. Paulette, R. G., C. S. Boruff, and J. 0. Nack. A pollution
abatement program for distillery wastes. Journal of the Water
Pollution Control Federation, 42:1387. (1970)
15. Vaughn, R. H., and G. L. Marsh. Disposal of California winery
wastes. Industrial and Engineering Chemistry, 45(12):2686.
(1953)
16. Vaughn, R. F., and G. L. Marsh. Problems in disposal of
California winery wastes. American Journal of Enolbgy, 7:26.
(1956)
17. Friedman, A. A. The Effect of Temperature on Activated Sludge
Growth Rate and Yield. Ph.D. Thesis, University of California,
Davis. (1970)
18. Methods for Chemical Analysis of Water and Wastes. U.S.
Environmental Protection Agency, Water Quality Office,
Washington, D.C. (1971)
19. Standard Methods for the Examination of Water and Wastewater,
13th edition. American Public Health Association, New York. (1971)
20. Hiser, L., and A. W. Busch. An 8-hour biological oxygen demand
test using mass culture aeration and COD. Journal of the Water
Pollution Control Federation, 36:505. (1964)
82
-------
21. Mullis, M. K., and E. D. Schroeder. A rapid biochemical oxygen
demand test suitable for operational control. Journal of the
Water Pollution Control Federation, 43:209. (1971)
22. Weissberger, A. Technique of Organic Chemistry, Vol. III.
Tnterscience Publishers, Inc., New York. (1956)
83
-------
SECTION VIII
PUBLICATIONS
Schroeder, E. D., D. J. Reardon, R. Matteoli, and W. H. Hdvey.
Biological treatment of winery still age. Presented at the
Fourth National Symposium on Food Processing Wastes, Syracuse,
New York. (March 26-28, 1973)
Matteoli, R., E. D. Schroeder, A. P. Jackman, and G. Tchobanoglous,
Physical treatment of winery still age. Presented at the 28th
Industrial Waste Conference, Purdue University, West Lafayette,
Indiana. (May 1-3, 1973)
84
-------
SECTION IX
GLOSSARY
ABBREVIATIONS
BOD
cm
COD
m
mg/L
mm
min
MLSS
RCF
RPM
Biochemical Oxygen Demand
centimeter
Chemical Oxygen Demand
meter
milligram per liter
millimeter
minute
Mixed liquor suspended solids
Relative centrifugal force
Revolutions per minute
SYMBOLS
g
kd
N
PH
r
rox
ss
-1 -2
gravitational constant ML t
-2
Maintenance energy coefficient, t
Rotational Speed, t"
Negative logrithm of the hydrogen ion concentration
Centrifuge radius, L
Specific organic removal rate, t
-1
Suspended solids concentration, ML
Cell yield
-3
Specific growth rate, t
-1
85
-------
SECTION XI
APPENDICES
Appendix Page
A 1971 Data 87
B 1972 Data 92
C Laboratory Reactor Data 98
D Solids Separation Data 112
86
-------
TABLE A-l. 1971 PILOT PLANT INFLUENT COD DATA
DATE
10/28
10/29
10/30
10/31
11/1
11/5
11/6
11/7
11/10
11/11
11/15
COD, mg/L
Raw Waste
Unfiltered
33,351
38,334
30,806
28,534
17,500
16,683
32,980
28,941
Filtered
14,721
16,263
18,657
16,971
16,473
15,557
15,409
15,594
14,844
Inlet Holding Tank
Unfiltered
34,401
31,197
30,904
35,150
34,458
16,499
30,759
12,598
Filtered
15,193
15,438
14,012
15,795
13,755
13,995
16,194
17,066
13,159
87
-------
TABLE A-2. 1971 ACTIVATED SLUDGE PILOT PLANT
DATE
10/25
10/29
10/30
10/31
11/1
11/5
11/6
11/7
11/10
11/11
11/15
COD, mg/L
Mixed Liquor
Unfiltered
14,170
15,575
16,776
18,703
16,932
17,500
14,574
18,136
20,660
13,395
Filtered
4,669
3,889
5,825
4,698
7,461
6,166
6,523
6,097
6,867
4,452
Effluent
Unfiltered
12,359
14,864
13,875
17,568
14,439
17,985
14,670
16,621
15,712
15,794
Filtered
4,985
4,478
5,825
7,073
6,166
6,220
5,598
6,665
4,452
88
-------
TABLE A* 3 1971 UNFILTERED ANAEROBIC PACKED BED COD DATA
DATE
4 f
£n •
10/25
i ^
10/29
'10/30
10/31
11/1
11/5
11/6
11/7
11/10
11/11
11/15
COD, mg/L
TANK 1
Bottom
56,043
67,407
48,723
60,947
82,598
77,712
65,113
90,666
Top
33,377
51,198
35,669
41 ,773
34,191
32,617
24,737
36,413
42,674
23,091
TANK 2
L_ Bottom
48,274
74,768
83,502
45,688
64,594
77,091
73,312
64,114
63,548
Top
29,700
35,339
44,417
38,721
34,613
41 ,462
44,390
47,980
TANK 3
Bottom
36,748
58,229
65,272
62,171
68,737
60,714
71,112
63,548
Top
15,360
34,164
35,458
34,472
40,304
33,711
28,288
33,081
49,036
30,788
89
-------
TABLE A-4. 1971 ANAEROBIC PACKED BED FILTRATE COD DATA
DATE
10/25
10/29
10/30
10/31
11/1
11/5
11/6
11/7
11/10
11/11
11/15
COD, mg/L
TANK 1
Bottom
17,476
18,505
18,620
17,065
16,012
16,376
16,695
18,093
17,066
18,601
Top
12,122
14,876
16,028
16,531
15,245
16,473
14,419
14,824
17,193
16,056
15,435
TANK 2
Bottom
17,161
17,028
17,206
19,392
16,683
16,570
15,936
16,823
17,193
16,460
16,721
Top
14,642
18,674
16,145
17,318
14,957
14,703
15,813
16,693
15,854
17,018
TANK 3
Bottom
15,114
18,701
18,620
21,331
16,695
17,631
17,893
17,873
19,491
Top
15,272
16,301
16,028
15,901
15,533
16,473
16,722
16,993
17,167
15,534
90
-------
TABLE A-5. 1971 TEMPERATURE AND pH DATA FOR AEROBIC AND ANAEROBIC SYSTEMS
Date
10/27
10/28
10/29
10/30
10/31
11/1
11/4
11/5
11/6
11/7
11/9
11/10
11/11
11/14
11/15
Inlet tank
PH
7.2
6.4
6.9
6.8
6.3
6.3
6.3
6.3
6.5
6.7
5.9
5.9
5.9
5.9
6.4
P°C
27
29
33
38
42
44
36
42
37
38
30
41
39
24
31
Anaerobic No. 1
pH T°C
Bot
6.2
6.2
6.3
6.3
6.2
6.3
6.2
6.4
6.3
5.8
6.4
6.0
5.9
6.3
6.2
Top
6.0
6.1
Bot
24
24
24
24
26
28
24
29
28
28
25
26
21
14
19
Top
26
29
Anaerobic No. 2
PH T°C
Bot
6.2
6.3
6.3
6.3
6.2
6.5
6.5
6.5
6.3
6.0
6.1
6.2
5.9
6.2
6.3
Top
6.1
6.1
Bot
19
22
23
20
23
25
25
24
24
25
22
24
24
13
17
Top
23
29
Anaerobic No. 3
pH T°C
Bot
6.2
6.2
6.3
6.3
6.2
6.2
6.3
6.3
6.3
5.8
6.0
6.0
6.0
6.2
6.2
Top
6.0
6.1
Bot
19
21
21
22
23
23
25
27
27
27
22
25
19
13
22
Top
24
29
Aerobic
pH TUC
7.7 21
6.9 12
7.2 16
7.6 21
7.5 21
7.4 24
7.4 21
7.5 24
7.4 22
7.0 19
7.3 22
7.3 27
7.1 22
7.4 13
7.3 17
-------
TABLE B-1. 1972 AEROBIC PILOT PLANT DATA
Date
9/22
9/16
9/18
9/19
9/20
9/21
9/24
9/25
9/27
9/28
9/29
9/30
10/1
10/2
10/5
10/8
10/10
Mixed
Liquor
PH
7.1
7.1
7.2
6.4
7.2
7.2
7.3
7.4
7.6
7.7
7.5
Influent
Susp.
Solids
mq/L
17,800
6,500
COD
mg/L
77,000
74,600
22,600
7,300
Filtrate
COD
mq/L
19,700
17,100
14,800
7,250
Effluent
Susp.
Solids
mq/L _
9,200
3,840
5,700
7,400
3,800
3,560
COD
mq/L
22,400
12,600
21 ,200
21 ,600
9,250
13,500
30,800
Filtrate
COD
mg/L '
>4,100
6,700
7,320
3,000
6,800
3,200
1,500
3,600
TABLE B-2. SEPTEMBER 24, 1972 ANAEROBIC
PACKED BED NUMBER 1 DATA
Sample
Port
1
4
6
COD
mq/L
133,000
7,500
9,300
Filtrate
COD
mq/L
12,800
700
7,500
Susp.
Solids
_mq/L
18,500
5,500
7,300
PH
5.5
5.9
92
-------
TABLE B-3. SEPTEMBER 24, 1972 ANAEROBIC
PACKED BED NUMBER 2 DATA
Sample
Port
1
4
6
COD
mg/L
54,800
8,900
27,300
Filtrate
COD
mg/L
13,200
3,900
Susp.
Solids
mg/L
21 ,400
7,100
4,100
PH
5.9
TABLE B-4. SEPTEMBER 12, 1972 ANAEROBIC
PACKED BED NUMBER 3 DATA
Sample
Port
1
4
6
COD
mg/L
66,500
40,000
12,400
Filtrate
COD
mg/L
28,700
28,500
6,500
PH
6.0
6.0
TABLE B-5. SEPTEMBER 24, 1.972 ANAEROBIC
PACKED BED NUMBER 3 DATA
Sample
Port
4
6
COD
mg/L
25,000
11,100
Filtrate
COD
mg/L
7,000
3,900
Susp.
Solids
mg/L
10,200
5,500
PH
5.8
5.9
93
-------
TABLE B-6. 1972 RAW STILLAGE SETTLING DATA*
September 17
Time
min.
0
7
16
18
19
22
23
25
27
29
31
35
38
49
58
77
87
107
123
174
1440
Interface
ht. , cm
280
274
232
224
210
199
193
182
175
168
162
153
144
123
109
i
90
September 19
Time
min.
0
13
22
30
40
70
84
Interface
ht . , cm
360
360
349
342
331
297
284
95 | 234
105
115
404
1440
83 ]
73 !
67
56
39
i 218
i
214
178
173
September 20
Time
min.
0
98
124
138
144
196
261
274
415
467
514
559
596
Interface
ht. , cm
280
266
258
253
252
235
212
207
171
162
154
151
148
* Samples run on September, 21 and October 12 did not settle.
94
-------
TABLE B-7. 1972 ACTIVATED SLUDGE
MIXED LIQUOR SETTLING DATA
September 20
Time
min.
0
149
201
248
293
330
1440
Interface
cm
280
95
87
53
50
48
39
September 21
Time
min.
0
5
22
35
45
51
103
226
255
288
353
1440
Interface
cm
280
274
272
263
255
252
221
168
157
146
95
50
September 23
Time
min.
0
10
27
45
52
75
90
123
150
200
227
268
1440
Interface
cm
280
277
273
266
262
251
244
227
207
179
168
151
56
95
-------
TABLE B-7. (cont.) 1972 ACTIVATED SLUDGE
MIXED LIQUOR SETTLING DATA
September 24
Time
min.
0
15
30
75
105
148
295
335
370
390
711
1440
Interface
cm
280
275
272
255
244
227
202
182
165
146
87
59
September 26
Time
min.
0
30
45
70
90
no
131
166
211
326
571
Interface
cm
280
277
274
266
253
244
232
202
174
154
112
September za
Time
min.
0
12
65
120
1440
Interface
cm
280
90
70
84
50
TABLE B-7. (cont.) 1972 ACTIVATED SLUDGE
MIXED LIQUOR SETTLING DATA
September 30
Time
min.
0
5
10
115
275
Interface
cm
280
62
45
39
34
October 2
Time
min.
0
30
60
75
155
180
250
320
450
1440
Interface
cm
280
272
255
252
218
207
188
176
129
84
October 3
Time
min.
0
50
275
350
560
1440
Interface
cm
280
274
244
235
202
84
96
-------
TABLE B-7. (cont.) 1972 ACTIVATED SLUDGE
MIXED LIQUOR SETTLING DATA
October 10
time
min.
0
25
75
110
175
200
410
590
1440
Interface
cm
280
272
252
238
212
202
140
118
84
October 11
Time
min.
0
20
40
85
145
205
240
460
900
1440
Interface
cm
280
266
249
221
193
174
165
134
109
84
October 12
Time
min.
0
20
80
140
320
580
1440
Interface
cm
280
274
249
227
179
140
90
97
-------
TABLE C-l. LABORATORY REACTOR 1 AT 6_ « 4.67 DAYS
v*
00
Date
6/13
6/15
6/16
6/17
6/18
6/20
6/22
6/23
6/26
6/27
Feed
COD
mg/L
18,318
18,047
18,675
18,328
Filtered
Feed COD
mg/L
16,963
14,472
16,531
16,363
Mixed Liquor
Filtrate COD
mg/L
794
573
696
620
713
601
Flow
Rate
mL/day
755
792
665
778
778
D.O.
mg/L
6.5
6.7
PH
6.2
6.3
Suspended
Solids
mg/L
7,820
7,490
7,225
6,960
6,775
-------
TABLE C-2. LABOTATORY REACTOR 2 AT 9,. = 4.67 DAYS
Date
6/15
6/17
6/18
6/20
6/22
6/23
6/26
6/27
Mixed Liquor
Filtrate COD
mg/L
890
794
716
838
761
792
435
D.O.
mg/L
7.5
7.8
pH
6.2
6.1
6.0
Suspended
Solids
mg/L
6,080
5,748
6,340
6,095
6,160
99
-------
TABLE C-3. LABORATORY REACTOR 1 AT 9C = 3.58 DAYS
CD
O
Date
7/1
7/2
7/3
7/4
7/5
7/6
7/7
Feed
COD
mg/L
17,621
19,365
19,003
19,739
19,396
19,739
Filtered
Feed COD
mg/L
15,636
16,598
16,916
17,384
17,186
16,977
Mixed Liquer
Filtrate COD
mg/L
2,892
544
515
503
566
558
Flow
Rate
mL/day
955
920
1,030
1,010
D.O.
mg/L
6.0
5.5
4.0
PH
6.4
6.3
6.1
5.8
Suspended
Solids
mg/L
6,740
7,000
7,468
-------
TABLE C-4. LABORATORY REACTOR 2 AT 9 =3.58 DAYS
Date
•
7/1
7/2
7/3
7/4
7/5
7/6
7/7
Mixed Liquor
Filtrate COD
mg/L
608
632
612
492
544
471
D.O.
mg/L
7.5
7.7
8.0
PH
6.1
6.0
5.9
5.6
Suspended
Solids
mq/L
5,743
5,775
6,240
101
-------
TABLE C-5. LABORATORY REACTOR 1 AT 6.. = 2.75 DAYS
o
ro
Date
7/9
7/10
7/n
7/12
7/13
7/14
7/16
7/18
7/19
Feed
COD
mg/L
20,335
19,641
19,641
20,448
19,965
19,904
Fi 1 tered
Feed COD
mg/L
17,656
17,388
17,143
18,162
17,806
17,960
Mixed Liquor
Filtrate COD
mg/L
648
643
615
651
638
665
Flow
Rate
mL/day
1,210
1,190
1,320
1,230
1,420
1 ,250
1,280
D.O.
mg/L
1.3
5.0
2.0
6.5
pH
6.1
6.1
6.1
6.0
Suspended
Solids
mg/L
8,503
8,918
8,930
8,785
8,535
-------
TABLE C-6. LABORATORY REACTOR 2 AT 6c = 2.75 DAYS
Date
7/10
7/11
7/12
7/13
7/14
7/16
7/18
7/19
Mixed Liquor
Filtrate COD
mq/L
702
626
588
570
534
606
D.O.
mg/L
8.0
7.5
7.4
7.5
PH
5.9
5.9
5.9
5.9
Suspended
Solids
mg/L
6,745
6,733
6,878
6,985
7,315
103
-------
TABLE C-7. LABORATORY REACTOR 1 AT 6 = 2.42 DAYS
Date
8/7
8/8
8/9
8/10
8/12
8/14
8/16
8/17
Feed
COD
mg/L
22,238
22,274
20,354
20,013
19,930
20,207
Filtered
Feed COD
mg/L
19,078
19,661
18,289
18,075
17,825
16,959
Mixed Liquor
Filtrate COD
mg/L
1,293
724
647
696
645
574
Flow
Rate
ml/day
1,480
1,440
1,440
1,460
1,410
D.O.
mg/L
c
o
CO
-------
TABLE C-8. LABORATORY REACTOR 2 AT 9
2.42 DAYS
Date
8/7
8/8
8/9
8/10
8/12
8/14
8/16
8/17
Mixed Liquor
Filtrate COD
mg/L
862
360
670
720
750
615
D.O.
mg/L
c
o
CQ
i.
O)
-M
O)
PH
6.1
6.1
6.1
6.1
Suspended
Solids
mg/L
6,508
6,907
6,758
6,813
7,550
6,723
105
-------
TABLE C-9. LABORATORY REACTOR 1 AT 6_ =
V*
1.87 DAYS
Date
8/20
8/21
8/22
8/23
8/24
8/25
8/26
8/28
8/29
8/30
Feed
COD
mg/L
14,922
23,777
20,236
19,004
18,679
Fi 1 tered
Feed COD
mg/L
13,707
21 ,752
19,710
17,500
16,755
Mixed Liquor
Filtrate COD
mg/L
525
623
628
593
565
Flow
mL/day
1,730
1,590
2,040
2,020
1,870
1,900
1,940
1,870
D.O.
mg/L
3.5
4.0
2.2
3.0
PH
6.0
6.1
6.5
6.4
6.7
6.3
Suspended
Solids
mg/L
7,980
9,478
(
8,685
8,685
8,515
-------
TABLE C-10. LABORATORY REACTOR 2 AT 6 =1.87 DAYS
c
Date
8/20
8/21
8/22
8/23
8/24
8/25
8/26
8/28
8/29
8/30
Mixed Liquor
Filtrate COD
mg/L
789
788
746
745
798
D.O.
mg/L
5.7
_ 6.3
7.0
6.5
PH
6.2
6.2
6.1
6.3
6.3
6.2
Suspended
Solids
mg/L
5,535
6,390
7,057
7,148
7,328
107
-------
TABLE C-ll. LABORATORY REACTOR 1 AT 6 = 1.41 DAYS
\f
CD
GO
Date
8/31
9/2
9/3
9/4
9/5
9/6
9/7
9/8
9/10
Feed
COD
mg/L
19,382
19,314
20,027
21 ,454
19,868
Fi 1 tered
Feed COD
mq/L
16,952
16,875
16,021
18,282
19,248
18,774
Mixed Liquor
Filtrate COD
mq/L
623
589
516
629
649
811
Flow
Rate
mL/day
2,470
2,520
2,470
D.O.
mg/L
3.5
3.5
2.3
PH
6.4
6.6
6.5
6.3
6.6
Suspended
Solids
mg/L
8,690
6,945
7,960
8,585
9,400
-------
TABLE C-12. LABORATORY REACTOR 2 AT 6c = 1.41 DAYS
Date
8/31
9/2
9/3
9/4
9/5
9/6
9/7
9/8
9/10 .
Mixed Liquor
Filtrate COD
mg/L
706
589
575
640
684
788
D.O.
mg/L
6.5
6.5
7.0
PH
6.2
6.3
6.5
6.3
6.6
Suspended
Solids
mg/L
7,713
6,580
6,455
6,542
6,460
109
-------
TABLE C-13. LABORATORY REACTOR 1 AT 6,. = 1.09 DAYS
Date
9/10
9/11
9/12
9/13
9/14
Feed
COD
mg/L
21 ,674
20,712
19,958
Filtered
Feed COD
mg/L
19,328
19,028
17,618
Mixed Liquor
Filtrate COD
mg/L
4,703
7,157
9,168
Flow
Rate
mL/day
3,240
3,190
D.O.
mg/L
1.3
3.8
1.0
1.6
pH
5.5
4.8
4.4
Suspended
Solids
mg/L
6,453
4,678
o
-------
TABLE C-14. LABORATORY REACTOR 2 AT 6
1.09 DAYS
Date
9/10
9/11
9/12
9/13
9/14
Mixed Liquor
Filtrate COD
mg/L
947
990
1,157
D.O.
mg/L
2.5
2.5
1.0
1.5
PH
7.2
7.5
7.5
Suspended
Solids
mq/L
7,850
6,870
111
-------
TABLE D-l. BATCH SETTLING TEST RESULTS FOR 1/3, 2/3 MIXTURE OF PURIFLOC
C-41 AND AMOCO CATIONIC COAGULANT AIDS*
Time
mln.
0
30
60
135
255
345
490
1440
Interface Height, cm
Coagulant Concentration, mg/L
0
280
269
255
196
140
123
109
92
7
280
276
271
260
244
230
207
134
15
280
276
271
267
246
232
211
136
24
280
276
270
251
210
182
150
107
36
280
276
271
241
176
146
120
101
60
280
274
266
238
176
146
121
99
* Still age COD = 42, 400 mg/L
TABLE D-2. EFFECT OF COAGULANT ADDITION ON SUPERNATANT LIQUOR*
Characteristic
COD, mg/L
Susp. solids mg/L
Coagulant Concentration mg/L
0
19,540
1,270
7
16,160
222
15
16,260
24
16,130
175
36
15,860
149
60
15,620
88
* Still age COD = 42, 400 mg/L
112
-------
TABLE D-3. EFFECT OF CENTRIFUGATION ON CAKE MOISTURE CENTRATE SUSPENDED SOLIDS* AND CENTRATE
COD**
Run
1
2
4
8
15
1200 RPM ^
Cake
Moist
%
90.6
88.0
87.6
87.3
87.1
Susp.
Solids
mg/L
1,685
1,255
1,003
872
COD
mg/L
19,800
18,870
18,120
18,290
17,550
1800 RPM
Cake
Moist
%
87.7
86.8
86.4
36.8
84.5
Susp.
Solids
mg/L
1,625
1,446
944
765
538
COD
mg/L
19,470
18,830
18,220
18,080
17,540
2400 RPM
Cake
Moist
%
87.9
88.7
85.5
84.4
83.7
Susp.
Solids
mg/L
1,147
968
609
648
501
COD
mg/L
18,630
18,220
17,720
18,110
17,540
3000 RPM
Cake
Moist
%
86.0
85.4
86.0
84.4
82.6
Susp.
Solids
mg/L
1,138
872
621
418
394
COD
mg/L
18,975
17,600
17,540
17,250
17,540
co
* Stillage Susp. Solids Cone. = 11,952
** Stillage COD cone. = 36,100
TABLE D-4. EFFECT OF POLYELETROLYTE ADDITION* ON CENTRATE CHARACTERISTICS AT 2400 RPM
Run
Time
Min.
0
1
3
6
10
15
Cake
Moist
%
-
85.8
83.9
84.5
83.8
81.8
Susp.
Solids
mg/L
23,010
1,610
897
644
506
598
COD
mg/L
46,400
21,105
21,120
20,955
20,715
* 8 mg/L Amoco Cationic
-------
TABLE D-5. EFFECT OF DETARTRATION ON SUPERNATANT COD*
CaC12
cone.
9/L
0
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.50
3.00
4.00
4.50
Ca(OH)2
cone.
9/L
0
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.50
3.00
4.00
4.50
pH
3.45
4.95
5.50
6.25
6.90
7.70
8.25
8.65
9.05
9.90
10.7
11.4
Filtered
COD
13,060
12,830
12,520
12,400
12,160
12,780
12,320
12,520
12,620
12,720
* Still age COD = 34,230
114
-------
TABLE D-6. SETTLING OF DETARTRATED STILLAGE USING PURIFLOC A-23*
1 mg/L
Time
min
0
20
40
50
80
140
170
200
230
250
1440
Interface
height
cm
280
279
272
261
262
249
241
234
227
221
138
10 mg/L
Time
min
0
10
20
30
40
50
60
70
80
100
110
120
130
1440
Interface
height
cm
263
256
246
236
225
207
196
187
182
174
167
163
157
123
14 mg/L
Time
min
0
10
20
30
40
50
60
70
80
100
no
120
130
1440
Interface
height
cm
268
260
247
241
230
220
211
202
192
184
178
171
168
129
20 mg/L
Time
min
0
10
20
30
40
50
60
70
80
90
105
125
135
1440
Interface
height
cm
263
259
254
248
241
235
229
221
213
207
196
183
178
129
* Untreated Detartrated Still age did not form interface.
115
-------
TABLE D-6. (continued) SETTLING OF DETARTRATED STILLAGE USING PURIFLOC
A-23
26 mg/L
Time
min
0
10
20
30
40
50
60
70
80
90
105
125
135
1440
Interface
height
cm
263
258
249
241
231
222
213
206
196
191
182
171
168
129
30 mg/L
Time
min
0
10
20
30
40
50
60
70
80
90
95
Interface
height
cm
274
241
204
185
-
162
157
151
146
145
144
40 mg/L
Time
min
0
10
20
30
40
50
60
70
80
90
Interface
height
cm
269
232
185
159
148
141
136
133
131
129
50 mg/L
Time
min
0
11
26
36
46
56
66
76
86
96
101
Interface
height
cm
272
221
165
151
142
138
136
132
131
131
131
116
-------
TABLE D-7. COAGULATION OF CENTRATE* WITH NALCO 610 POLYELECTROLYTE
AND BENTONITE.
Nalco 610
mg/L
5
15
35
60
100
Supernatant Liquor COD, mg/L
10 mg/L Bentonite
17,840
17,770
17,070
16,480
16,410
20 mg/L Bentonite
17,070
16,510
16,200
15,900
15,610
*Stillage COD = 39,100 mg/L
Centrate COD = 19,080 mg/L
Fitrate COD = 17,470 mg/L
117
-------
TABLE D-8. DETARTRATION OF CENTRATE*
Ca(OH)2
q/L
0.00
0.25
0.50
0.75
1.00
1.30
1.50
1.75
2.00
2.25
2.50
2.75
3.00
4.00
CaC12
q/L
0.00
0.25
0.50
0.75
1.00
1.30
1.50
1.75
2.00
2.25
2.50
2.75
3.00
4.00
COD
mq/L
18,960
16,240
16,080
16,120
15,690
15,930
15,760
15,230
15,340
14,990
14,750
14,790
14,630
13,500
pH
3.60
4.10
4.31
4.72
5.60
6.40
6.98
8.55
9.10
9.60
10.05
10.05
10.65
11.5
TABLE D-9. COAGULATION OF DETARTRATED
CENTRATE WITH PURIFLOC A-23
Pun" floe
A-23
mg/L
0
2.5
5.0
10
25
35
50
75
Residual Cone
mq/L
Suspended
Solids
5820
2872
2430
2020
1480
1210
1002
819
COD
15,040
13,260
13,840
12,520
11,900
11,450
11,700
10,800
118
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-660/2-75-002
3. RECIPIENT'S ACCESSION-NO.
TITLE AND SUBTITLE
5. REPORT DATE
February 1975
Pilot Scale Treatment of Wine Still age
6. PERFORMING ORGANIZATION CODE
AUTHORiS)
8. PERFORMING ORGANIZATION REPORT NO.
E. P. Schroeder
PERFORMING ORG MXIIZATION NAME AND ADDRESS
California Department of Agriculture
Wine Advisory Board
717 Market Street
San Francisco, CA 94103
10. PROGRAM ELEMENT NO.
1BB037
11. CONTRACT/GRANT NO.
12060 HPC
12, SPONSORING AGENCY NAME AND ADDRESS
Pacific NW Environmental Research Laboratory
National Environmental Research Center
200 SW 35th Street
Corval 1 is, Oregon
13. TYPE OF REPORT AND PERIOD COVERED
Final - Aug. 1971-Ann. 1Q73
14. SPONSORING
Aug. 1971-Aug.
NG AGENCY CODE"
16. SUPPLEMENTARY NOTES
18. ABSTRACT
Pilot and laboratory scale studies were run on aerobic and anaerobic biological treat-
ment of winery stillage over a two year period. The pilot scale studies included work
with aerobic lagoons and anaerobic packed towers. Laboratory systems studied were
aerobic reactors without recycle and batch fed anaerobic processes. Because
suspended solids removal proved to be a key factor in successful biological treatment,
centrifugation, detartration, coagulation and flocculation, and combinations of these
methods were included in the studies.
Centrifugation proved to be the best method of removing solids prior to biological
treatment. Solids removal in combination with an aerobic treatment process can be
expected to produce final filtrate chemical oxygen demands of about 700 mg/L and a
final filtrate BOD of about 75 mg/L. Anaerobic processes studied did not operate
well but produced effluents with chemical oxygen demands of the order of 4000 mg/L.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
* Industrial Wastes, * Waste Identifi-
cation, Winery Waste Water, Biological
Treatment, Solids Separation
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/GrOUp
* Waste Water Treatment,
Agricultural Wastes
13/13B
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
125
20. SECURITY CLASS (This page)
22. PRICE
Form 2220-1 (9-73)
U.S. GOVERNMENT PRINTING OFFICE: 1975-698-091/10* REGION 10
------- |