.STUDY OF RUM DISTILLERY WASTE
TREATMENT AND BY-PRODUCT
RECOVERY TECHNOLOGIES
SCS Engineers
Long Beach, 'Cal ifornia 908.07
Contract Number 68-03-2573
Project Officers
Harry Thron
Kenneth Dostal
Donald Wilson
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
-------
CONTENTS
Page
Figures i i "
Tables iv
I. Conclusions 1-1
II. Recommendations II-l
III. Introduction III-l
IV. General Approach IV-1
V. Disposal to a Public Owned Treatment Works V-l
(POTW)
VI. Ocean Disposal VI-1
VII. Land Disposal VII-1
VIII. Evaporation VIII-1
IX. Pollutant Reduction- through Process IX-1
Modifications
X. Aquaculture X-l
XI. Biological Treatment Systems XI-1
XII. Bibliography XII-1
Appendices
A. Aquaculture Correspondence A-l
B. Anamet Plants 8-1
-------
FIGURES
Number Page
VI- 1 Proposed ocean outfall. San Juan, Puerto VI- 8
Rico (Bacardi)
VI- 2 Proposed outfalls for Puerto Rico Distillers, VI-10
Arecibo
VI- 3 Proposed outfall for VIRTU'" ... ' . ' VI-12
VII- 1 Artist's conception of land cultivation VII-27
si te
VII- 2 Representative site profile VII-35
VIII- 1 Potassium Salt Recovery from Crude Potash VIII-32
VIII- 2 Process flow diagram for the production, of
org'anic compound fertilizer
IX- 4 Molasses centri fugat.ion IX- 1
IX- 2 Yeast recycling by centrifugation of IX- 2
fermenter beer
IX- 3 Removal of fermenter bottoms and yeast IX- 9
XI- 1 Typical Anamet System XI-10
XI- 2 Yeast Cultivation in Molasses Distillery XI-17
Stillage
11
-------
TABLES
Number Page
VII- 1 Composition of Selected Constituents in VII-13
Mostos on a Dry-Weight Basis
VII- 2 Annual Quantities of Mostos Generated by VII-14
Bacardi, Puerto Rico Distillers, and VIRIL
VII- 3 Estimated Annual Mostos and Irrigation VII-16
Water Application Rates for Bacardi
(Tultitlan , Mexico) and Serralles
(Mercedita, Puerto Rico)
VII- 4 Estimation of Maximum Allowable Heavy Metals VII-18
and Sludge Loadings for Agriculture Crops
and Mostos Loadings for Land Farming
VII- 5 Estimated Site Life Based on Annual VII-20
Application Rate of 12 ac-in/yr
VII- 6 Minimum Estimated Land Requirements for VII-24
Bacardi, Puerto Rico Distillers and VIRIL
as a Function of Application Rates
VII- 7 Basis for Conceptual Design VII-26
VII- 8 Conceptual Design Waste and Site Parameters VII-34
VII- 9 Conceptual Design Vehicle Requirements for VII-37
.. Land Farming Utilizing Tanker Trucks for
Mostos Transportation
VII-10 Labor Requirements for Land Farming Utilizing VII-37
Tanker Trucks for Mostos Transportation
VII-11 Labor Requirements for Land Farming Utilizing VII-38
a Pipeline for Mostos Transportation
VII-12 Capitalists for Three Conceptual land VII-39
Farming Sites Utilizing Tanker Trucks for
Mostos Transportation
IV
-------
TABLES (continued)
Number Page
VII-13 Capital Costs for Three Conceptual Land VII-42
Farming Sites Utilizing a Pipeline for
Mostos Transportation
VII-14 Annual Operation and Maintenance Costs for VII-44
Three Larid Farming Sites Utilizing Tanker
Trucks for Mostos Transportation
VII-15 Annual Operation and Maintenance Costs for VII-45
Three Land Farming Sites Utilizing a
Pipeline for Mostos Transportation
VII-16 Summary of Annual Capital and Operating VII-46
Total Costs for Tanker Truck vs. Pipeline
Mostos Transportation
VII-17 Annual Land Disposal Costs of Mostos as a VII-43
Function of Rum Production
VIII- 1 Mean Chemical Composition of Cane CMS VIII-18
VIII- Z Amino Acid Analysis of Cane CMS . VIII-19
VIII- 3 ,-. Estimated Maximum Annual Quantities of CMS VIII-22
Which Could be Incorporated into Puerto
Rican Mixed Animal Feed'
VIII- 4 Estimated Annual Potassium Production as an VIII-34
Incineration By-Product
VIII- 5 Puerto Rican Annual Commercial Consumption VIII-35
of Potash and Potassium Compounds in
Direct Application Materials
IX- 1 Cost Estimates for Pollutant Reduction IX-V1
Through Process Modifications, from Puerto
Distillers and VIRIL
IX- 2 Breakdown of Cost Estimates for Pollutant IX-12*
Reduction Through Process Modifications,
from Puerto Rico Disti11ers and VIRIL
/*
XI-12 Recommended Operating Parameters for XI-12
Anaerobic Digestion in the Anamet System
-------
I. CONCLUSIONS
GENERAL
Initial investigations showed the following four technolo-
gies are not applicable for the. treatment of mostos:
- Coagulation and sedimentation
- Foam separation
- High gradient magnetic separation
- Reverse osmosis
In-depth .investigation showed that the following technolo
gies are also not applicable:
- Discharge of mostos to a POTW (on Puerto Rico and St.
Croix)
- Aquaculture
Land farming, the Anamet process, evaporation, inci-
neration and organic fertilizer production, appear to be
feasible technologies. Their direct application to the
treatment of mostos, however, has not been fully demon-
strated.
Agricultural utilization of mostos on sugar cane as cur
rently practiced by Serralles Oestileria in Mercedita,
^
Puerto Rico, and Bacardi in Tultitlan, Mexico, appears
to be a viable disposal alternative.
1-1
-------
Aerobic biological processes such as activated sludge
*
and trickling filters are not applicable when considered
solely by themselves.
0 Anaerobic digestion is partially effective in reducing
BOD and suspended solids.
Pilot plant tests of the Anamet and evaporation processes
at Puerto Rico Distillers and Bacardi, respectively, had
not been finished prior to the completion of this study.
Consequently, conclusions could not be developed with
respect to their individual performance.
" The economics of mostos treatment and disposal technology
are far from being clearly defined. The present practice
of discharging untreated mostos into the ocean through
short outfall pipes is inexpensive. All other treat-
ment or disposal alternatives investigated are by con-
trast very costly. For example, estimates for land farm-
ing ranged from approximately 8 to 16 cents per proof gal
Ion produced. The cost of.treatment and disposal by the
evaporation (CMS) production process, the Anamet treat-
ment process, and incineration were not determined. It..
is hoped that the pilot plant research presently under
way will provide accurate cost ranges.
TECHNOLOGY SPECIFIC
Land Disposal
- Land reclamation does not appear to be a viable dis-
posal option
1-2
-------
- Agricultural utilization appears to be a practical
disposal option for Puerto Rico Distillers.
- The combination of potentially available land and dry
climate, suggests that VIRIL is suitably located to
utilize land farming as a disposal option
- Bacardi is less favorably situated and may not be able
.to locate, sufficient land near their Catano facility
to make transport of mostos to the disposal site eco-
nomically feasible.
Evaporation
A review of available literature indicates that
evaporation of mostos to condensed molasses solubles
(CMS) is a technologically feasible treatment alternative
The utilization or ultimate disposal of CMS, however, may
not be economically practical.
Estimates of local (Puerto Ricanl CMS consumption
suggest that approximately 18 percent of the total theo-
retical CMS production could be utilized as an animal
feed supplement if incorporated in the feed at the five
percent, by weight, level. '.
Incineration
The recovery of potassium salts from incinerator asr.h
is estimated to result in an annual supply of 5,550 T of
t»
<20, which is approximately five times greater than the
current (1965-1975) average annual consumption of direct
1-3
-------
application materials. The total potassium market is
uncertain.
Organic fertilizer production
Utilizing CMS as a raw material, an annual estimated
53,500 metric tons of an organic N-P-K fertilizer could
be produced by the three distillers. The annual local
consumption of this type of fertilizer was not available
from the literature reviewed.
Biological treatment processes
Available data suggests that, while neither aerobic
or anaerobic processes are, in themselves, satisfactory
treatment alternatives, the combination of the two pre-
sent technologically feasible treatment options. One
such process known as Anamet, has been successfully
utilized on waste streams similar but not identical to
mostos. This suggests that other similar combined bio-
logical systems, while not specifically investigated
during this study, may have applications to the treatment
of mostos.
1-4
-------
II. RECOMMENDATIONS
Results of the Anamet and CMS pilot plant tests should
be incorporated into any future evaluation of rum wastes.
A program should.be initiated to evaluate the potential
for land farming of mostos. Pilot studies should be con-
ducted in Puerto Rico and St. Croix to determine:
- Degree of mostos pH adjustment required
- Compatibility of land farming techniques with site
specific topographical and hydrological conditions
- Disposal application rates as a function of climate,
methods, of application, so'lid.s loading, etc...
- Degree of mostos degradation as a function of time
- Accumulation of mostos constituents within the soil
matrix and the extent of cationic and anionic migration
through the 'soil column
- Potential aesthetic problems.
The feasibility of mostos discharge by Puerto Rico Dis-
tillers to nearby government-owned sugar cane fields
should be investigated.
,
The Bacardi Corporation should consider a fusel oil and
"heads" recovery program, thus eliminating this compo-
nent from the mostos stream.
II-l
-------
Recent innovative aerobic biological treatment processes
should be investigated for potential use following
anaerobic digestion. Included among these are:
- UNOX, a proprietary pure oxygen activated sludge pro-
cess developed by Union Carbide
- Deep Shaft, developed by Eco-Research
- Carrousel, developed by Envirotech.
The potential for local (Puerto Rican) and export con-
sumption of CMS should be fullly investigated.
A study investigating the short and long-term markets
for local organic fertilizers consumption should be
initiated.
A study should be initiated to determine the total
potassium market for Puerto Rico and St. Croix.
II-2
-------
III. INTRODUCTION
On December 6, 1977, the United States House of Representa-
tives presented a mandate to the United States Environmental Pro-
tection Agency (EPA) to study the compatibility of pollutants dis-
charged from Puerto Rico and Virgin- Islands rum distillers with
the marine environment. Specifically, the EPA was to determine:
The environmental acceptability of rum distillation
wastes
The potential benefits or adverse effects of marine
di sposal
« The geographical, hydro!ogtcal , and biological charac-
teristics of the marine receiving waters around Puerto
Rico and the Virgin Islands
Waste processing technologies which facilitate the
recovery of nutrients and/or other beneficial materials
in the wastes
. Treatment technologies which reduce the discharge of rum
distillation waste pollutants.
In 1973, the EPA initiated a study program aimed at fully
t»
assessing the opti.ons for disposal of rum distillery wastes and
associated environmental consequences. The primary objectives
of this study are (1) to identify and assess various treatment
III-l
-------
technologies and by-product recovery options available to the rum
distillers, and (2) to indicate the existence of information
voids so that research and development efforts can be directed
toward the filling of these gaps.
III-2
-------
IV. GENERAL APPROACH
Selection of treatment and by-product technologies for ini-
tial investigation was the first project objective. An initial
list of treatment and by-product recovery technologies was pre-
pared from a variety of sources, including:
In-house EPA information on previous and related work
Interviews with personnel from The Bacardi Corporation,
Puerto Rico Distillers, and The Virgin Islands Rum Indus-
tries, Limited (VIRIL)
Intensive literature review of available foreign and
United States research projects dealing with the treat-
ment of cane molasses distillation residues and related
wastes and processes
Unpublished and in-press data made available by the rum
di s.ti 11 ers
Current university research projects
Interviews and correspondence with knowledgeable consul-
tants, and industrial equipment manufacturing represen-
tatives.
/*
This assemblage of information and data resulted in the
following list of potential treatment and by-product recovery
technologies for preliminary evaluation.
IV-1
-------
Treatment of wastes by a public owned treatment works
(POTW)
Biological treatment processes
- Aerobic
Trick!i ng f i1ter
Activated sludge
- Anaerobic
Anaerobic digestion
- Anaerobic/aerobic systems
Coscul1euela
t Anamet
- Other
Cultivation of fodder yeasts, bacteria, and fungi
Evaporation
- Production of condensed molasses solubles (CMS) for
animal feed
- Utilization of CMS as an intermediate product
Incineration with conventional ash disposal
Incineration with potassium recovery
Organic fertilizer production
Land disposal
-Agricultural utilization
- Land reclamation
/
- Land farming
Process modifications
- Molasses pretreatment
IV-2
-------
- Removal of fermenter beer
- Recovery of fusel oils and heads
- Removal and recovery of yeast from fermenter beer
Ocean disposal
Aquaculture
Physical/chemical treatment processes
- Coagulation and sedimentation
- Foam
- High- gra.dfent magnetic separation
- Reverse osmosi s.
SELECTION PROCESS
While each of the above treatment and by-product recovery
technologies may be considered as a potentially viable alternative
to the current waste discharge problems of the rum distillers,
some of the alternatives have greater application potential than
others. Most of the technologies have never been applied to rum
wastes., although they have been' successfully employed for the
treatment of similar high strength fermentation wastes. Fur-
thermore, the time constraints of this investigative effort dic-
tated that the major research efforts be directed toward only
those promising technologies. With these factors in mind, the
following criteria were adopted to evaluate each alternative:
.*>
Extent and ,substance of available literature
f* " ~
9 Direct application experience of the technology to the
treatment of rum distillery wastes
IV-3
-------
Direct application experience of the technology to the
treatment of similar wastes - (e.g. beet molasses fer-
mentation residuals)
Performance and operating characteristics of the treat-
ment process on high strength waste streams
t Estimated investment and annual O&M costs
Energy consumption
0 Potential for retrofitting technology to existing plant
facilities
0 Ultimate disposal of primary and secondary wastes - (e.g
final effluent, sludge, ash).
Other considerations also entered into the evaluation pro-
cess. For example, the fact that the distillers are located on
islands considerably removed from the contiguous states and the
limited availability of land and other resources were important
factors.
Each technology was qualitatively reviewed in terms of the
above criteria and either accepted or rejected for more inten-
sive investigation.
TREATMENT TECHNOLOGIES AND BY-PRODUCT RECOVERY OPTIONS
CONSIDERED UNFEASIBLE
The technologies considered unfeasible are:
0 Sedimentation and coagulation
0 Foam separation
0 High gradient magnetic separation
0 Reverse osmosis.
In the following section, an explanation is given for the
rejection of each of these technologies.
IV-4
-------
Sedimentation and Coagulation
Sedimentation for treating rum distillery waste was found
to be unsatisfactory by Biaggi (Jl)-due to the fact that almost
all of the waste solids were in solution. Siaggi (11) and Sen
et al . (.92) reported no success with the chemical coagulation
process a.s a means- of treating rum waste. Even with extremely
high dosages of ferric chloride, aluminum sulfate, lime, ben-
tonite, and various other coagulants, no floe formation was
observed.
Foam Separation
Foams may be used to remove substances dispersed in a liquid
by several processes known collectively as foam separation- (83).
The common feature of all these processes is that separation is
brought about by virtue of differences in the surface activity
of the dispersed substances. In contrast to most other treat-
ment processes, foam separations work most efficiently with dilute
solutions.
Air or other gases bubbling through the liquid generate gas-
liquid interfaces and result in foaming, which allows the inter-
faces to be collected. Surface-active substances are adsorbed
onto the rising bubbles and concentrate in the foam. The sub-
stances then can be removed from the liquid.
Foam techno! ojj'i es applicable to the treatment of mostos
/
include froth flotation for the removal of suspended solids,
and foam fractionation for the removal of certain ionic species
(metals , etc) .
IV-5
-------
Certain problems and shortcomings are inherent to the use
of foam separation for the treatment of mostos. These are:
t- Foam separation technologies do not remove the soluble
BOO from the waste stream. Since the principal compo-
nents of mostos are soluble solids, further treatment of
the waste stream would be required.
Aerating mostos in a laboratory scale activated sludge
study has resulted in excessive foaming (11). The addi-
tion of gas for foam separation would probably increase
the foaming problem.
No treatment studies have been completed on the applica-
tion of foam separation technologies to mostos. No trans
fer technology exists for the foam treatment of similar
wastes.
High Gradient Magnetic Separation (H6MS)
Magnetic separation techniques have- been used since the
nineteenth century to remove tramp iron and to concentrate iron
ores. A variety of conventional magnetic separation devices are
in wide use today. These devices generally separate relatively
coarse particles of highly magnetic material containing large ;
amounts of iron from non-magnetic media.
In recent years, magnetic devices have been developed which. :>
/
are capable of separating even weakly magnetic materials on
the order of one micron in size at high waste stream flow rates.
These so-call.ed "hig-h gradient magnetic separators" have been
IV-6
-------
designed to maximize the magnetic forces on fine, paramagnetic
materials. The separations may be economically operated at pro-
2
cess rates of up to several hundred gpm/ft (!)-
High gradient magnetic separators typically consist of a
canister type device packed with fibrous ferro-magnetic material.
They are magnetized by a strong external magnetic field sur-
rounding the canister. Magnetic particles contained in fluids
flushe.d through the canister are trapped on the edges of the
magnetized fibers ; non-magnetic particles pass through unaffected,
This type of system has been demonstrated to work effectively in
treatment of wastewaters containing low levels of impurities (1).
Magnetic components can be separated directly from waste
Streams, whereas non-magnetic contaminants (organics, colloids,
etc.) must be given a "magnetic handle" by seeding the. stream'
with finely divided magnetic seed material. Binding to the seed
is accomplished by adsorption or coagulation.
Adsorption mechanisms are not fully understood, but are pos-
tulated to include chelation or surface bonding. The more
practical technique is chemical coagulation of the contaminants
into aggregates.
The potential for HGMS as an option for treating mostos
appears to be unfavorable (US'.). The system relies heavily on
the ability of alum-^or other flocculants to produce aggregate
particulates. It has been demonstrated by Biaggi (11) and Sen
et al. (92), however, the sedimentation was ineffective in treat-
ing mostos due to the high percentage of soluble constituents in
IV-7
-------
the waste stream. Further, the HGMS system seems favorable as
a secondary or tertiary treatment module, handling waste streams
whose soluble and particulate constituents are in relatively low
concentrations. Mostos, on the other hand, has a soluble solids
content of approximately 10 percent.
Reverse Osmosis
Reverse osmosis (RO) is a process for removal of dissolved
solids in water and wastewater. The name is derived from the
fact that, in the process, water is made to flow from a solution
of higher concentration through a semi-permeable membrane to a
solution of lower concentration, the opposite of natural osmosis.
To cause this flow, a pressure greater than the osmotic pressure
of the water or wastewater being treated must be applied. In-
practice, pressure considerably in excess of the osmotic pres-
sure (.400 psi is a common feed pressure in desalination plants)
is required to obtain a water flux through the membrane,which is
sufficiently large to make the process effective and practical
(25', 26, 70).
The majority of research and development on RO to date has
been concentrated on applications involving the treatment of
brackish waters. However, some investigation of the applicabil-
ity of RO to the treatment of municipal wastewater has also ':
occurred. In addit4on, the cost of power and membrane replace-
ment is high.
Pilot programs of RO treatment of wastewater have been con-
ducted under EPA sponsorship.
IV-8
-------
In these studies it was found that substantial pretreatment
was required prior to operation of the RO unit to remove dis-
solved organics, which otherwise fouled membranes. Pretreatment
included chemical flocculation, coagulation, and filtration plus
activated carbon adsorption. These constitute a very expensive
pretreatment chain (25, 26, 98)!.. Additionally, RO technology
has. been generally applied to waste streams in which total dis-
solved solids (TDS) were not in excess of 7,000 ppm. Mostos, on
the other hand, has a TDS concentration of 77,,400 to 85,600 ppm
(93). Therefore, RO could not be implemented without some, form
of dilution, an option which is not realistic, especially so on
St. Croix, an island which has suffered under drought conditions
over the last several years. Therefore, reverse osmosis is not
considered to be a feasible treatment technology for. rum distil-
lery wastas.
FEASIBLE TECHNOLOGIES
The following chapters present an analysis of each treatment
technology and by-product recovery option. Throughout the text,
rum distillery wastes have been identified by one or more of
several synonyms. These are: rum distiller's wastes, rum slops,
slops, and mostos. They are all defined as the combined liquid
discharge resulting from fermentation of sugar cane molasses.
IV-9
-------
V. DISPOSAL TO A PUBL-IC OWNED
TREATMENT WORKS (POTW)
INTRODUCTION
Manufacturing facilities located within any of the 48 con-
tiguous states and faced with one of several wastewater treatment
optio.ns generally have a POTW- in reasonably close proximity. If
all options, prove technologically OP economically unfeasible,
the POTW can be selected as a final choice. Decision parameters
such as present and future surcharges, peak flows, 300 equivalents,
increased or decreased plant production must necessarily be fac-
tored into the final selection of treatment alternatives. The rum
industry on both Puerto Rico and St. Croix., the U.S. Virgin
Islands, should, therefore, be examined with discharge to a POTW
as a viable- di sposaT option.
The following discussion presents such an examination on a
site-specific basis.
Bacardi
The Bacardi Corporation is located approximately 1 1/2 mi
northwest of Catano. It is bounded on the north by Ensenada de
:j:v
Boca Vieja and by the small village of Palo Seco, and to the east
i-
by San Juan Say. 'The overall region lacks an adequate sewage
system. This inadequacy presently causes gross pollution to the
waters of the region from domestic and industrial wastewater
sources (1). The Puerto Rico Aqueduct and Sewer Authority (PRASA)
V-l"
-------
has approved the Bayamon Wastewater System to provide sewage ser-
vices to areas of Aguas Buenas, Catano, Toa Baja, Toa Alta, and
San Juan. The wastewater system to be constructed will consist
of trunk and branch sewers to collect wastewater from local sewer
systems in existing and projected population centers through the
year 2020. Included in this system is a treatment plant (POTW)
designed to accomodate the regions expected wastewater. The POTW
will be located in Barrio Palmas east of the Bayamon river channel
in the municipality of Catano about 1 km from the sea coast of
Ensenada de Boca Vieja. The plant will be constructed in 10 MGD
modules and in treatment phases.
The initial phase will be designed for primary treatment with
average capacity of 40 MGD; effluent will flow directly to the
ocean through an ocean outfall to be constructed simultaneously
with the trunk and. POTW systems (5). Construction has not been
implemented on either the POTW,.or the ocean outfall. The future
date for upgrading the POTW to secondary treatment is now known.
Previous analytical investigations of mostos reported an
average total dissolved solids content of 7.77 percent, represent-
ing approximately 93 percent of the total solids present (93). Fur-
ther, mostos is extremely high in soluble BOD and COD, with res-
pective average ranges of 26,500 - 47,400 ppm and 72,000 - 92,00.0
ppm. The technological considerations of primary treatment are '
i*
such that they wo,uld not be expected to remove any significant
quantity of either. If discharged to the POTW, the Bacardi efflu-
ent would be diluted and discharged essentially untreated to the
ocean. Therefore, the alternative of discharging to a POTW for
V-2T,
-------
the purposes of pollutant reduction is effectively not available
to Bacardi",
Puerto R'ico Distillers
Puerto Rico Distillers is located on the western edge of
Arecibo, a city which currently discharges its domestic and indus-
trial streams directly to the ocean. Construction has reportedly
begun on a trunk line system to route domestic and industrial
wastes to a. primary POTW (also under construction) located on
the eastern side of Arecibo @7). There are reportedly no pre-
sent plans to install a trunk line to the western extremities of
the city - the location of the distillery. Design capacity,
estimated completion data, future timetable for upgrading to
secondary treatment,, etc.,, were not available in the literature
reviewed and from personal correspondence." Further, since the
mostos discharged from Puerto Rico Distillers is virtually iden-
tical to that of Bacardi, removal of any significant quantities
of pollutants would not be expected. Therefore, discharge to a
POTW for Puerto Rico Distillers is not a viable option.
Virgin Islands Rum Industries, Limited (VIRTU
The VIRIL plant is located in a rural area about 5 mi east
of Frederiksted, St. Croix. A small POTW with primary treatment
is located several miles east of the distillery, on the eastern
edge of the airport property. Recently, a secondary module was **
constructed several hundred yards from the primary facility and
is currently in operation. The POTW serves several newly developed
housing areas and the airport facilities. Current average flow
is about 0.5 MGO; design capacity is 4 MGD (113). Secondary effluent
V-3--
-------
is discharged to the ocean through a 48 in diameter, 9,000 ft
outfall (71).
VIRIL and POTW personnel have conducted preliminary discus-
sions concerning.the possibility of pumping via pipeline, the
mostos and otherplant wastestreams to the POTW. The POTW person-
nel suggested that considering the high strength and waste volume
(population equivalent is approximately 90,000), the only accep-
table method of receiving mostos would be in a diluted state so
that the BOO would not exceed that of domestic sewage (200 - 300
ppm BOD) (71). Fresh water, sufficient for the necessary dilution,
however, is not available on this water-starved island.
Dilution with seawater is not possible due to the toxic
effects of sodium chloride on POTW microorganisms. Therefore,
discharge to the POTW is not a viable alternative to VIRIL.
SUMMARY '
Discharge of mostos to a POTW is not a viable or practical
alternative for any of the rum distillers in Puerto Rico and St.
Croix, U.S. Virgin Islands
V-4,
-------
VI. OCEAN DISPOSAL
Outfall pipes into the ocean are intended to return contami-
nated fluids to the environment in a way that promotes adequate
transport and dispersion of the waste fluids. Urban and industrial
communities who discharge their complex mix of wastes to the
ocean are typically located along the shallow edge of the ocean,
with often tens or even hundreds of kilometers of contaminated
shelf between the shoreline and the deep ocean. The bottom slope
on the shelf is typically less than one percent. Thus, outfalls
extending several kilometers offshore discharge into a body of
water of large latera.l extent compared to the depth.
The decay times-of the wastes are important in the choice of
effective discharge strategies. For example-, the problems of
very persistent contaminants such as DOT cannot be alleviated by
dispersion from an outfall. Such wastes must be contained at the
source and prevented from entering the environment. On the other
hand, degradable organic wastes may be effectively disposed of .
through a we! 1 -desi gned ocean outfall. Since the typical decay .-.
/
time is only a few days, potential problems are only local and '&
»
not regional or global.
Coastal waters are biologically productive because of the
infusion of nutrients from nearby land masses and upwelling of
nutrient-rich bottom water. Man's additions of organic nutrient
VI -1
-------
material cause perturbations in the coastal ecosystems. Most
past problems have resulted from excessive waste loadings very
close to shore or in estuaries with poor circulation patterns,
that is, in cases where man's effects have been overwhelming.
The current practice of outfall design for large systems is to
build long pipelines (>3 k.m) on the open coast to depths up to
70 m, and to install large multiple-port diffusers. They can
achieve large initial dilutions of the order of 100 to 1 and
produce submergence of the entire wastewater cloud below the
surface when there is sufficient ambient density stratification.
This practice has greatly improved local water quality at rela-
ti vely 1ow costs (55).
A typical outfall consists of a submarine pipeline roughly
perpendicular to the shoreline. In shallower water.it is buried-
under rip-rap for wave protection and, at greater depths,-it is
laid on the ocean floor with ballast rock on either side. At the
far end, there is usually a diffuser section, a manifold with
many small holes to distribute the flow throughout a large ocean
area.
Effective utilization of the ocean for assimilation of waste-
water depends on both engineering and environmental considerations
Engineering feasibility for ocean diffusion depends upon good jet;
mixing and oceanic convection and diffusion. Important factors
are ocean currents-, density stratification, waste decay constants,
and the use of multiple-jet diffusers.
VI-2
-------
SUMMARY OF CONSIDERATIONS FOR AN OCEAN OUTFALL
Initial planning for an ocean outfall should include an
oceanographic survey in the vicinity of possible discharge sites
to determine (16):
1. Currents (direction, magnitude, frequency, variation
with depth, relation to tides, water displacements)
2. Densities (variation with depth determined from salinity .
and temperature data and standard tables)
3. Submarine topography, geology, and bottom materials
4-. Marine biology
5. Turbidity
6.. Dissolved oxygen,, etc..
The final site selection for an ocean outfall is usually
based on general characteristics of the coastal waters and on
topography of the drainage area. Details of diffuser design are
developed after the general site is chosen.
DESIGN FOR DIFFUSION
Since the objective of an ocean outfall is to disperse waste
effluent in the ocean, first consideration is given to the analysis
of diffusion. From such an analysis one establishes a satisfactory
length for the outfall from the shore, a desirable diffusor pipe":
arrangement and the approximate number and spacing of ports. ..'
There are essentially three stages of diffusion or turbulent"
/*
mixing of wastes discharged into the ocean (16):
1. Initial jet mixing (considering jet strength, currents,
and density differences)
2. Development of a homogeneous diffusion field
-------
3. Turbulent diffusion of waste field as a whole due to
natural oceanic turbulence.
An excellent treatment on the above mentioned stages of
diffusion is presented in papers by N. H. Brooks (16, 17, and 18).
Some of the highlights of these papers are discussed here.
Port Design
The outlet ports may quite satisfactorily be circular holes
in the side of the pipe without nozzles or tubes or other pro-
jecting fittings. For optimum dilution, the jets should discharge
*
horizontally, with no initial upward velocity. The inside of the
port should be. bell-mouthed to minimize clogging and to provide
a discharge coefficient which will remain constant over a period
of years.
Number of Ports
Research has shown that a multiple-outlet diffuser greatly
enhances the possibility of generating a submerged waste field
(T<6). if the waste is discharged at a single port or "en masse,"
its dispersion and dilution will be slower than if it is dis-
charged over a large area through a number of ports. In fact,
without the use of multiple-outlet diffusers, other conditions
being equal, much longer outfalls into deeper water are necessary
to provide the same degree of dispersion. In general, the greater
the depth (volume), the greater the dispersion.
/
Velocity in Diffus»er
The flow velocity in all parts of the diffuser should be high
enough to prevent deposition of any residual particles. For
settled sewage, velocities of 2 to 3 feet per second at peak flow
VI-4!
-------
are adequate (but borderline). If deposition takes place in any
part of the diffuser over an extended period of time, the cross
section of the pipe may become so constricted that locally the
-velocity will be reduced, a cycle that will accelerate the in-pipe
deposition and clogging process...".: "
Flow Distribution
The. outflow between the various diffuser ports should be
fairly uniform'. If the diffuser is laid on a sloping sea bottom,
it will be impossible to achieve uniform distribution between
ports for all flow rates. In such cases, it is desirable to
,
make the distribution fairly uniform at low or medium flow, and
let the deeper ports discharge more than the average port dis-
charge during high rates of flow.
If waste, pumping is necessary or the available gravity head
is limited, the total head loss.in any diffuser should be kept
reasonably small. Additional head losses of a few feet are
usually adequate.
Prevention of Seawater Intrusion
All ports should flow full in order to prevent the intrusion
of seawater into the pipe. Seawater entering the pipe will be
stagnant and will tend to trap grit and other settleable matter,
reducing the hydraulic capacity of the diffuser.
SITE SPECIFIC CONSIDERATION
,»
The fol 1 owing--section presents ocean disposal options
available to each rum distiller predicated on the following
assumptions:
1. The chemical and physical characteristics of mostos are
such that ocean disposal results in minimal environmental
VI-3-
-------
harm to local ocean flora and fauna.
2. Practice of removing "heads" and "fusel oils" is con-
tinued by Puerto Rico Distillers and VIRIL and is
adopted by Bacardi .
Bacardi Corporation
The Bacardi Corporation manufacturing facility (Figure VI-1).
is located just south of the town of Palo Seco, on a peninsula of
land bounded on the east by San Juan Bay and on the west by both
the Palo Seco Thermoelectric Plant and the Bayamon River. Approxi-
mately one mile southeast of the plant is the. town of Catano.
The Bayamon River currently receives its water from the Thermo-
electric Plant (approximately 100 MGO - cooling water) and from
Bacardi (approximately 0.2 MGD - condensate, boiler blow down,-'
cooling water, fermenter washings, etc.)
Design has been submitted and approved for the construction
of the Bayamon Wastewater Treatment Plant (initial average flow
of 27 MGO -primary treatment only), and an ocean outfall accepting
the combined flow of the proposed Bayamon and existing Puerto
Nuevo Treatment Plants. The combined effluent will be discharged
through a 120-in-diameter concrete pipe extending 4,000 ft into \
the Atlantic Ocean to a depth of about 160 ft below mean ocean ':
sea level. The wasf.ewater will -be diffused to the ocean through
p.orts located along two 1,000-ft diffusers. Initial dilution
is estimated to be about 150-175:1. The outfall pipe (Figure VI-1)
will be located on the far eastern edge of the Bacardi property.
VI-6
-------
. If Bacardi were to utilize this newly designed system, two
options would be available:
Option 1 : Discharge to the Bayamon Wastewatsr Treatment Plant
Bacardi would discharge its combined effluent (approximately
0.5 MGO) to the proposed primary treatment facility. The
Bacardi effluent would be combined with other domestic and
industrial wastes, thus affecting dilution, and would be
discharged to the ocean. However, the nature of mostos. is
such that the high 300 and COO loads are a function of the
high percentage of solubilized organics (nonfermentable
sugars, acids, etc.) and would not be appreciably reduced by
primary treatment. Hence, the mostos would essentially pass
through the primary system unchanged.
Option 2: Discharge directly to the outfall.
Initial flow through the outfall pipe is anticipated to be
40 MGD. By running pipe across the plant, grounds and
connecting directly to the outfall pipe, the 0.3 MGD Bacardi
mostos stream would receive an instantaneous dilution of
130:1. Assuming that the diffuser operates within its
design limits, the Bacardi effluent will be further diluted
*
upon discharge to the ocean. The overall mostos dilution is;
estimated to be 20,000:1.
The difference between each of the above alternatives is
f
basically one of semantics. Both dilute the waste to the same
degree with final discharge to the ocean. However, in the first
case, discharge is technically to a POTW; in the latter, discharge
is directly to the ocean without benefit of "treatment".
VI-7
-------
Figure VI-1.
Proposed ocean outfall
( Bacardi ).
San Juan, Puerto Rico
VI-8
-------
Puerto Rico Distillers
Puerto Rico Distillers is located in the western section of
Arecibo, bounded on its eastern, western, and southern borders
by encroaching city development and to the north, by the Atlantic
Ocean (Figure VI-2). Two options are presented.
Option 1: 1,600-ft outfall.
A 6-in-diam'eter concrete (C = 130) pipe will extend 1,600 ft
into the Atlantic Ocean to a depth of approximately TOO ft.
Based on a volume of 0.4 MGD (280 gal/min), effluent would
flow at a velocity of about 3.2 ft/sec. Head loss was cal-
culated 'to. be 7.4 ft/1,000 ft or 11.8 ft for the entire
1,600-ft outfall. Pumping requirements are estimated at
1.1 brake horsepower (BHP).'
Option 2: 2,000-ft outfall.
A 6-in-diame.ter concrete (C - 130) pipe will extent 2,000 ft
into the Atlantic Ocean to a depth of approximately 150 ft.
Based on a volume of 0.4 MGD (280 gal/min), effluent would
flow at a velocity of 3.2 ft/sec. Total head loss would be
14.8 ft. Pumping requirement is estimated at 1.4 BHP.
Initial dilution for either option is estimated at 50:1.
If the density of mostos could be lowered to less than that of
seawa.ter (for example, by diluting mostos with other lower
VI -9
-------
., ( 156 FT DEEP)
Figure V'I-2. Proposed outfalls for Puerto Rico Distillers, Arecibo.
-------
specific gravity waste streams), mixing- would be expected to
increase. If the outfall is extended to 3,000 ft, the waste
stream would be discharged at a depth of approximately 400 ft.
Virgin Islands Rum Industries, Limited (VIRIL)
The VIRIL plant is located in the southwestern area of
St. Croix, approximately 1-1/2 mi east of Long Point. Speci-
fically, the plant is on the Diamond Estate of Princess Quarters
approximately 1-1/2 mi north of the ocean. The mostos dis-
charge pi pe,. original ly a sewer outfall, runs from the plant,
across the estate known as Betty's. Hope, and terminates approxi-
mately 1,000' ft in the ocean at a depth of about 15 ft (see
Figure VI-3).. .It is worthy of note that the ocea'n slopes very
gradually throughout this area so that any outfall pipe designed
for depths of TOO ft is reached, the degree of slope increases
rapidly. For this reason, only one option has been examined.
Option 1: 150-ft outfall.
An outfall to a depth of approximately 150 ft. Approxi-
mately 22,000 ft of 4-in-diameter concrete pipe would be
required. VIRIL's estimated 140-gal/min effluent would
-flow at 3.5 ft/sec. Head loss is estimated at 13.6 ft/1,000
ft or 300 ft for the entire net fall. Pumping requirements,
would be equivalent to 14.2 BHP.
Due to the rapid slope of the ocean bottom at the outfall
/*
termination point^and the density limitations of the waste stream,
diffusion ports would probably not be required. Rather, the waste
stream could be permitted to exit directly from the end of the
VI-TT
-------
OUTFALL
(32,000 FT FROM PLANT)
Figure VI-3. Proposed outfall for VIRIL.
-------
outfall. Dye studies would have to be performed to investigate
the velocity and direction of the ocean currents to ensure that
the stream would not be diffused into the shallower areas.
The continuation of mostos disposal to the ocean will be
largely dependent upon the results of marine biological studies
currently in progress. Consequently, a complete economic analysis,
including capital and O&M costs for a pipeline, fs premature.
However, to place the economics of disposal in proper perspective
to other disposal technologies, pipeline estimates are presented
for Puerto Rico Distillers and VIRIL. It is assumed that costs
include pipe, connectors, ballast and/or- cement (for anchoring
pipeline to ocean floor), and labor for installation. Further,
it is assumed that, the above-mentioned costs for a 6-in-diameter
pipe will be approximately $125/ft and for a 4-i n-diameter plpe,.
$100/ft (46). Extending these unit costs to the- pipeline lengths
shown in the preceeding section gives the following:
« Puerto Rico Distillers
Option 1 (1 ,600-ft outfall - 6-in-diameter pipeline)
1 ,600 x $125 = $200,000
Option 2 (2,000-ft outfall - 6-in-diameter pipe)
2,000 x $125 = $250,000
VIRIL
Option 1 (22,000-ft outfall - 4-in-diameter pipe)
/'22,000 x $100/ft = $2,200,000
Costs for Bacardi were not estimated because they would be
utilizing the proposed POTVI outfall.
VI-13
-------
VII. LAND DISPOSAL
INTRODUCTION
Soil, with its tremendous surface area and varied microbial
population, has been viewed as a physical , chemical , and biologi-
cal filter for waste materials (98).. The waste is retained by
the soil matrix, and the organic fraction of the waste is even-
tually decomposed by the soil microorganisms [79,93).
Land application is a practice whereby wastes are applied to
the soil surface and incorporated into the top 6 to 12 in of soil.
As in farming practices, only the surface soil is utilized, and
the process of waste assimilation is aerobic. Optimal degradation
of the added wastes requires environmental conditions that are
similar to those needed for raising a crop: adequate amounts of
sunlight, water, nutrients, and oxygen. In addition, land appli-
cation and farming employ similar types of- equipment.
Wastes may be applied to soils for a variety of reasons.
Principal among these are the following:
1. Agricultural Utilization - as a soil amendment, utilizing
the available organic and inorganic nutrients to enhance^
the growth-of certain crops.
»
2. Land Reclamation - as a soil amendment to add organic and
inorganic nutrients to distressed or submarginal lands.
V11 -1
-------
3. Land Farming - as a means to dispose o'f wastes on a
dedicated piece of land, whereby, the organic consti-
tuents are degraded by natural processes (oxidation and
reduction via sunlight) and the inorganics are trapped
within the surface soil matrix.
The objective of this investigation is to explore the con-
cept of land disposal of mostos and to discuss in detail, that
particular method of land application which is technologically
and economically most feasible. As will become apparent in the
following pages, land farming for all three distillers studied
and agricultural utilization for Puerto Rico Distillers, were
selected to offer the most feasible land disposal options. Agri-
cultural utilization for Bacardi and VIRIL and land reclamation
for all three were excluded from more serious considerations for
the following reasons:
Agricultural utilization - from the literature reviewed,
research relating the effects of mostos on crop growth
could not be found. However, two other rum distilling
facilities, Serral 1 es .in'.Mercedi ta , Puerto Rico and
Bacardi in Tultitlan, Mexico have successfully disposed
of mostos by distribution (after dilution with fresh irri-
gation water) through canals irrigating nearby sugar- >.
cane fields. In this instance the feasibility of dis-
/
tillery discharge depends upon the availability of nearby
sugar cane fields. Puerto Rico Distillers is the only dis-
tillery in a similar situation in such...a favorable locale.
VI1-2
-------
Agricultural utilization of mostos to other crops cul-
tivated in the islands is dependent on significant crop
land availability and the adequate research expressing
'the effects of mostos on such 'crops (affinity to absorb
heavy metals, organics (toxic or otherwise), pesticides,
etc....]. Currently, neither of these are available. The
reader is referred to the section, Estimation of Mosto
Application Rates, for a more detailed explanation of the
Serralles and Bacardi, Mexico land disposal operations.
Land Reclamation - land reclamation programs are typically
employed in a situation where a waste generator is in
close proximity to sub-marginal or distressed land, e.g.,
areas that have been heavily mined. These circumstances
are not known to exist within any practical distance of
three; di stil 1 eries.
LITERATURE REVIEW
From the literature reviewed, the only research that could
be found presenting relationships between mostos and soils was
that of Raul Perez Escolar (75-78). A series of four articles
resulting from his studies was published in the Journal of Agri-
culture of the University of Puerto Rico in.1966 and 1967. Sum-
marized, they are as follows: >
.₯£
"Reclamation of a Saline-Sodic Soil by Use of Molasses
/»
and 0 i s t i-Tl ery Slops," (July 1966) - An application of
2.3 ac-in of mostos to a saline-sodic Faith clay of south-
western Puerto Rico was shown to lower both the soil con-
ductivity and exchangeable sodium. Further, he demonstrated
VII-3
-------
that the soil was reclaimed, resulting in an excellent
stable soil structure (75).
0 "Use of Molasses and Distillery Slops with Sulfur for the
Reclamation of a Saline-Sodic and a Sodic Soil from Puerto
Rico," (January 1967) - Tests and results similar to the
above cited references with sulfur included to enhance
reclamation. Testing was expanded to include both a
Guanic clay and a Faith clay (78).
"Stability of Soil Aggregates Treated with Distillery
Slops or Blackstrap Molasses," (July 1966) - Mostos,
applied at a minimum of 1/4 ac-in, was shown to produce
a long lasting and favorable effect on soil aggregate
stabilization which persisted after the growth of four
" consecutive crops (77).
"Separation and Characterization of the Active Soil-
Aggregating Agent Present in Distillery Slops," (October
1967) - The soil aggregating and stabilizing agents were
found to be present in the 80 percent alcohol insoluble
fraction of mostos with a composition of a mannose-
bearing polysaccharide (6 percent), protein (7 percent),
and a caramel (87 ' percent) (76).
References to mostos decomposition mechanisms and kinetics ''
were not avai 1 abl e/rom the materials reviewed. Further, referen-
ces to land disposal systems could not be found in this litera-
ture. However, several personal communications(33, 103, 104)
revealed the following land application programs:
VII-4
-------
Serralles - Mercedita, Puerto Rico
The Serralle's rum distillery is located in southwestern
Puerto Rico in Mercedita, several miles east of Ponce.
It is situated on approximately 4,000 acres, all of which
is owned by the Serralles family, except for some acreage
on which a sugar mill is located. Annual production of
rum is approximately 4 million proof gallons. Estimated
mostos flow is approximately 100,000-150,000 gal/day.
Since 1935, mostos has been discharged through a sys.tem
of canals which supply irrigation water to about 400 acres
of sugarcane. "The mostos is combined with other plant
waste's, achieving final dilutions ranging from 1/2 [other
plant wastes/mostos) to 5/1 (104). This combined stream
is pumped to a central irrigation station which further
dilutes the wastes prior to final field discharge.. The
combined plant wastes at this point are diluted approxi-
mately twofold. Irrigation water is supplied by an adja-
cent river originating in the mountains north of the plant.
In 1975, a system of peripheral ditches was constructed
to prevent mostos runoff during periods of heavy rain.
Soil and groundwater monitoring programs have not been
established at those sites although, no deleterious effects
have been reported for either the soil or sugarcane (103,
104). -
Bacardi, Tultitlan, Mexico - Bacardi operates a rum dis-
tillery in Tultitlan, Mexico (near Mexico City). Five
VII-5
-------
hundred acres of sugarcane are located approximate!y one
mile from the plant. There is no sugar mill present on
the site, which would provide processing water for further
mostos dilution. The mosto stream is combined with other
plant wastes achieving a dilution of 5/1 (.other plant
wastes/mostos). The combined solution is mixed with fresh
irrigation water further diluting the mostos resulting
in a final concentration ranging from 11/1 - 19/1. The
mixture, in turn, is distributed throughout the fields
(via canals and ditches) simultaneously supplying water to
the sugarcane.- crop and disposing of the mostos. This pro-
cess has reportedly been in operation for 17 yr, the only
apparent problems being occasional ponding, resulting in
some foul odors. The native soil is alkaline and the
acidic nature of mostos is reported to aid in maintaining
near neutral soil conditions (33).
Other Mexican Operations - Other distilleries in Mexico
reportedly discharge their rum wastes in a similar manner
but are able to achieve much greater dilutions as a result
of added sugar mill processing water. Mostos dilutions
at these operations are reported to range from 100-500
to 1 .
COMPONENTS OF A LAN.D DISPOSAL SYSTEM
,*
The implementation of a successful land disposal program is
dependent upon the interrelationships of many variables including
the selection of a disposal site, the nature of the wastes being
VII-6
-------
disposed, method of waste application, and economics. More speci
fically, these considerations involve the following parameters:
Selection of Disposal Site
- Land area required
- Land avai1abi11ty and cost
- Proximity to distilleries
- Proximity to homes and commercial establishments
- Ready access from all-weather roads
-Climate (rainfall, freedom from flooding, prevailing
wind direction)
- Field slopes and general topography
- Hydrogeology (location of groundwater)
- Soil characteristics
- Types of crops historically planted
- Ultimate land use of disposal site.
t Waste Characteristics
- Quantities of waste available
- Final waste disposal form [liquid, concentrate)
- Physical, chemical, biological properties
- Variability of constituents.
Application of Wastes
- Methods of transportation (irrigation pipe, tanker,
rail car,- barge . .)
/*
- Methods^of application (spreading, spraying, ridge and
furrow irrigation, subsurface injection, etc...)
- Hydraulic loading limitations
V11 - 7
-------
- Metals loading, pH, nutrient constraints
- Existing local and federal regulations
- Monitoring requirements.
Economics
- Capital costs (land, equipment, installation...)
- Operation and maintenance costs
- Overall disposal costs (total and factored).
SITE DESCRIPTIONS
Puerto Rico
Puerto Rico has an area of approximately 3,435 sq mi. The
island has three principal physiographic divisions. The most
extensive consists of the complex mountain ranges, which include
nearly all of the interior of the island. The less extensive are
the playa1. pi ains and the coastal plains, which circle the outer
edge of the island in a ribbonlike band. The elevation ranges
from sea level to 4,398 ft above, and the relief ranges from level
to precipitous. Ecologically, the complex mountain ranges are
of volcanic origin, and the coastal plains and playa plains are
sedimentary. The principal rocks include several types of Ter-
tiary and Cretaceous limestones shales, volcanic ash, lava, granite,
andesite, sandstone, serpentine, tuff, and conglomerate (86). There
are 358 soil types, phases, and miscellaneous land types recognized
on the island, representing a total of 115 soil series (86).
The populatio'n dens i ties, wh i ch vary greatly from place to
place, create the most serious social and economic problems con-
fronting thegovernment. A high proporation of the island area con-
sists of shallow, stony, steep, or submarginal land (86).
VII-8
-------
The present population is estimated to be approximately 3.4
million (1976-3.2 million). Projections to the year 2000, suggest
a population as high as 5 million (46). Available land, therefore,
is anticipated to be of premium value and reserved for the growth
of crops critical to the welfare of the people and the government.
The climate is nearly ideal for a 12-month growing season.
It is tropical, uniform, and oceanic, and nearly ideal conditions
exist for a heavy precipitation over all but the southwestern and
the extreme northwestern parts of the island, The average annual
precipitation ranges from less than 30 in in the southwestern
part to nearly 200 in on the high peaks in the Sierra de Luquillo.
Most of the cultivated crops are irrigated in districts receiving
lass than 45 in of mean annual rainfall, and sugarcane and grape-
fruit are irrigated in districts receiving less than 60 in. The
climate is nearly ideal for the production of such crops as coffee,
sugarcane, pineapples, bananas, mangoes, coconuts, yautias, and
yuca, but it is poor for alfalfa and berries. Such crops as wheat,
oats, barley, apples, pears, and peaches are not grown. Rice is
well adapted to the climate and to many of the soils, and will
probably be cultivated to a much greater extent in the future (86).
Rainfall in the San Juan and Arecibo areas has averaged about
60 in/yr, evaporation, about 82 in/yr. >..
Bacardi-- ^*
,»
The Bacardi Corporation is located approximately 1 1/2 mi
northwest of Catano. It is bounded on the north by Ensenada de
Boca Vieja and the village of Palo Seco; on the east by San
VII-9
-------
Juan Bay, and on the immediate west by the Palo Seco Thermoelec-
tric Plant. Further west is the small village of Levitown. To
the south are some marsh lands reserved for industrial park devel-
opment and town expansions(5) . Therefore, the only direction ,
available for Bacardi to land dispose their wastes is west. Some
acreage once used for sugarcane- production is approximately 4-5
mi west of the plant, but information as to the amount of land,
ownership, cost, future planned use, etc., was not available from
the sources contacted.
Puerto Rico Distillers--
Puerto Rico Distillers is located on the western edge of
Arecibo. The plant is bounded on the north by the Atlantic Ocean,
and on the east and south by the encroaching fringes of the city.
Sugarcane fields, owned by the government are located approxi-
mately 10 mi west of the plant. The planned use, ownership, etc.,
of land between the sugarcane fields and the plant was not avail-
able from the sources contacted.
St. Croix, U.S. Virgin Islands
The U.S. Virgin Islands include the islands of St. Thomas,
St. John and St. Croix. The last is the largest of the three with
an area of 84 sq mi. "
The topography is somewhat different from the other two with
a broad expanse of -low, relatively flat land running along the
,* ,
south 2/3 of the island. A range of hills, ranging in eleva-
ation from about 500 ft to more than 1,000 ft, topped by Mount
Eagle at 1,165 ft, runs along the northern coast. In the eastern
VII-10
-------
end of St. Croix is found another group of slightly lower hills
with a maximum elevation of about 860 ft. The relatively small
area covered by hills on St. Croix results in rather steep slopes
down to the Carribbean in the north and to the level areas to the
south. St. Croix is the only U.S. Virgin Island with any sizable
expanse of flatland suitable for farming. Here, sugarcane, which
was the principal crop, has been abandoned (84). Subsistence
crops are now a minor effort. Some cattle are raised for milk
and meat (84).
In St. Croix, industrial growth has become a significant fac-
tor in the island's economy (84). With the downgrading of agricul-
ture,, industrial complexes have been expanded to include the petro-
chemical industry and refinement o-f aluminum.
One. of the principal causes of concern in the U.S. Virgin
Islands is the short supply of water. Rainfall, while averaging
about 40 in annually (rainfall has been well below average over
the past several years) over most of the area, is insufficient.
This is due partially to a high evaporation rate (average about
73 in/yr) and rapid runoff.
In an effort to utilize available water efficiently, most
homes and business establishments catch rainwater on the roofs
and pipe it to cisterns. Generally, during the drier portion of^
the year, it is necessary to carry water by barge from Puerto
*
Rico.
Virgin Island Rum Industries, Limited (VIRIL)--
The VIRIL plant is located about 5 mi east of Frsderiksted,
St. Croix. Unlike Bacardi and Puerto Rico Distillers, VIRIL is
VII-11
-------
situated in an area which is just beginning to show signs of
development. The 25-ac plant is in a rural area surrounded by
400 to 500 ac of uncultivated land (not owned by VIRIL). Some
evidence of urban expansion, however, can be seen. To the south
and east about 1/2 mi away, several small housing developments
have recently been constructed, and .to the north, approximately
1/2 mile, ground has been dedicated to the construction of a hos-
pital. The area south of the plan extending to the Carribbean
(approximately 1 mi) is not inhabited and is covered with brush,
trees, and natural grasses.
WASTE CHARACTERISTICS
In the processing of rum, wastewater streams are generated
from four or five sources. Mostos has been previously identified
as the stream representing approximately 66 percent of the flow,.
and 98' percent of the pollutants(93) . For the sake of the follow-
ing presentation of data, it is assumed that only the mostos stream
will be land disposed.
Table VII-1 presents the chemical character!sties of mostos
but on a dry weight basis. It should be noted that approximately
69 percent of the dry solids are in the form of dissolved vola-
tiles- namely organics. While the exact nature of these are
uncertain, they are most likely a mixture of sugars, organic acids,
amino acids, protei,ns, polysaccharides, and various organic salts.
All are decomposable but the kinetics and degradation times are
not known. Total Kjeldahl nitrogen is 1.36 percent (dry wt basis),
indicating that suff i cient ..nutr i ents are probably present to
VII-12
-------
TABLE VI.I-1. COMPOSITION OF SELECTED CONSTITUENTS
IN MOSTOS ON A DRY-WEIGHT BASIS.
Parameter
Total solids
Total dissolved
volatile solids
Total Kjeldahl nitrogen
Cal cium
Potassium
Chloride
Sulfate
Zinc
Cadmium
Lead
Copper
PH
Previously
Reported*
8.35
5,79'
0.114
0.209
0.426
0.211
0.412
9.89
0.18
1.10
32.8
4.36
Expressed on Dry-
Wei'ght BTa's'i's'
I
69.3
1.36
2.50
5.10
2.53
4.93
Tig/£
118
2,16
13.2
393
V 11 -1 3
-------
sustain a reasonable level of " bacteriological activity.
Heavy metals are in the ranges typically reported for munici-
pal sewage sludge - another waste often disposed to land. On
the other hand, calcium, potassium, chlorides and sulfates are
extremely high and would probably have adverse effects on germi-
nation of seeds and subsequent growth of certain crops.
Table VII-2 presents estimated annual quantities of mostos
generated by each distillery
ESTIMATION OF MOSTOS APPLICATION RATES
Agricultural Utilization
As previously mentioned, both Bacardi (Tultitlan, Mexico)
and Serralles (Mercedita, Puerto Rico) have active and long-
termed agricultural utilization programs which involve irrigation
canal distribution of mosto through sugarcane fields. Both of
these plants have one thing in common - they have sugarcane fields
TABLE VII-2. ANNUAL QUANTITIES OF MOSTOS GENERATED BY
BACARDI, PUERTO RICO DISTILLERS, AND VIRIL
Average Daily
Discharge
Distiller (gals)
Bacardi Corp. 300,000
Puerto Rico 200,000
Distillers
VIRIL 100,000
Total
Annual Total Annual
Average Number Generation Dry Solids
of Operating of Mostos Generated*
Days (gals) . (Tons)
300 90,000,000 33,066
200 40,000,000 14,696
200 20,000,000 7,346
150,000,000 55,110
*Based on 8.35 percent solids
VII-14
-------
in close proximity to their distilleries. Of the three distiller-
ies to which this study is addressed, only Puerto Rico Distillers
is located in such an area. It is reported that approximately
4,000 or more acres of government-owned sugarcane fields are within
a five-mi radius of Puerto Rico Distillers (66, 103, 104). . If
arrangements could be made with the- owner-operators of these
fields, it may fae possible to transport (via. pipeline) the entire
plant waste stream to these fields where, upon further dilution
with fresh water, the combined streams could be used for irrigation
of the sugarcane.
From existing data on both the Bacardi (.Tultitlan) and
Serralles disposal operations, estimates of mosto application rates
(dry weight basis) and irrigation rates were made (33, 103, 104).
These are presented in Table. VII-3. Assuming that mostos is equally
distributed over the available acreage, the mosto application rate
is equivalent to 20.2 and 28.7 dry T/ac/yr at Bacardi and Serralle's
respectively. Similarly, from the information available, annual
irrigation application rates of the diluted mosto streams were
calculated to be 30.7 ac-in for Bacardi and 10.2 ac-in for Serralles.
The amount of irrigation water supplied by natural rainfall is
approximately 30 inches at Serralles, but was not available for
Bacardi.
It would seem, therefore, that Puerto Rico Distillers, with '
,+
approximately 4,00*0 acres of sugarcane in reasonably close proxi-
mity may have a favorable disposal alternative not available to
the others. Further, Puerto Rico Distillers' annual mostos dis-
charge is estimated to beapproximate!y 40,000,000 gals - about
V11 -1 5
-------
i
_j
en
TABLE VII-3. ESTIMATED ANNUAL MOSTOS AND IRRIGATION WATER APPLICATION RATES FOR
BACARDI (TULTITLAN, MEXICO) AND SERRALLES (MERCEDITA, PUERTO RICO)
Plant
Bacardi
Serralles
'i
Estimated Annual Mostos
Production Estimated Fresh Water:
(million gal) Muslos Ratio
27.5 IS:>
31.3 3.5:1
Sugar Cane
Acres
500
400
Estimated Annual Mustos
Loading Rate!
(T/ac) (dry ueiglit basis)
20.2
20.7
Estimated Annual Irrigation
Water Supplied to Dispose
of Hostos (ac-ln/yr)
30.7
10.2
Based on 250 operating days/yr.
Assuming inostos with specific gravity of 1.06 (8.B Ib/yal).
-------
20 to 30 percent higher than Serralles or Bacardi, Tultitlan.
The greater amount of sugarcane acreage available to Puerto Rico
Distillers suggests that tnostos could be discharged to the fields
on an even lower dry weight basis than is currently estimated for
Serralles and Bacardi, Tultitlan.
LAND FARMING "
There are two principal considerations which will determine
the maximum waste application rate for a land farming operation:
t Quantity of heavy metals in the mostos applied
Hydraulic capa.city of the site (the ability of the soils
to absorb moisture in addition to the annual precipitation)
A third consideration, land requirements, will determine the
practicality and feasibility of a land farming program.
Heavy Metals--.
While, there are no criteria governing the application of mos-
tos to soil, the presence of heavy metals suggests a similarity
to municipal sewage sludge, for which there are recommended limi-
tations. In most situations mostos will be applied to soils in a.
manner analagous to sludge. Therefore, regulations and practices
which control the use of sludge will be applied throughout this
analysis. Table VII-4 presents the U.S. Department of Agriculture
(USOA) recommendations for the total amounts of four metals con-;i
i'&
tained in sludge, wjiich may be added to agricultural soils (54)."
The calculations -in column four were derived from the following
formula :
<-*, rriv.u r/*~\ - Net, max allowable metal loading (Ib/ac)
mostos (dry T/ac) - ppm meta1 in mostos (dry wt) x 0.002
VII-17
-------
TABLE VI1-4. ESTIMATION OF MAXIMUM ALLOWABLE HEAVY METALS AND SLUDGE LOADINGS FOR AGRICULTURE
CROPS AND MOSTOS LOADINGS FOR LAND FARMING
I
'oo
Metal
Lead
Zinc
Copper
Cadmium
USDA Maximum
Allowable
Metal Loading*
i
2,000
1,000
500
20
Estimated
Amount of
Metal in Soil
at the Site
Ibs/ac
25
100
20
1
Maximum Allowable
Metal Loading
for
Agricultural
Soils
1,975
900
480
19
Maximum Allowable
Mostos Loading
for
Agricultural
Soils
74,811
3,814
611
4,318
Maximum Allowable
Mostos Loading
for
Land Farmingt
Tons/ac
374,055
19,070
3,055
21,590
*Total amount of waste metals allowed on agricultural land for soils with CEC>15 meg/1OOg (54).
(Based on preliminary EPA estimate that approximately 5 times the agricultural soil metal
accumulation would be the maximum acceptable loading for non-agricultural soils.
-------
The calculation assumes that the concentrations of these metals
(refer to Table VII-1) will remain constant over the life of the
disposal site. As shown in Table VII-4, the maximum allowable
mostos loadings vary considerably with respect to the individual
metals, with copper being the. limiting consti tutent.
For land farming purposes, no recommended criteria are avail-
able. Conversations with EPA personnel indicate that loading
rates as high as 5 times USDA criteria for agricultural soils may
be acceptable for non-agricultural soils if the metals applied
ar'e demonstrated, to be retained in the plow zone. If a landfarm
site will not be utilized for agricultural purposes at any future
time, the estimated maximum allowable waste loading in column
6 of Table VII-4 may be applicable. In this case, the'heavy metal
concentration of the. waste is not a significant constraining fac-
tor in determining the annual application rate to the. site if good
soil management is practiced. Soil pH is of specific concern since
.mostos is particularly acidic (pH - 4-.4). 'In order to minimize
the potential for metal.migration under acidic soil conditions,
mostos may have to be neutralized to maintain a soil pH of 6.5 -
7.5. However, the. accumulation of metals in the soil will limit
the site life. The life span of the site with respect to each
metal can be calculated as follows:
i ,- f- /1/r\ _ Maximum waste loading (T/ac)
^J ' ~ Annual application rate (T/ac/yr)
Table VII-5 presents these data. For mostos, the site life
would be.limited by copper accumulation, although a site life of
VII-19
-------
TABLE VII-5. ESTIMATED SITE LIFE BASED ON ANNUAL
APPLICATION RATE OF 12 AC-IN/YR*
Element
Lead
Zinc
Copper
Cadmium
Al 1 owabl e
Agricul tural
Site
625
32
5
36
Years of Disposal
Non-Agricul tural
Site
3,125
160
25
180
*Mostos assumed to have a specific gravity of 1.06 with a total
solids content of 8.35 percent. Annual application of mostos
equivalent to 120 dry Tons/acre.
25 yr is not a significant constraint. Note that the calculation
is based on the assumptions that the annual application rate
remains the same (12 ac-in/yr at 8.35 percent total solids) and
that there is no change in metal concentrations in the waste gener-
ated. An annual application rate of 12 ac-in/yr was selected
because it represents an equivalent loading of 120 dry Tons/ac.
For comparisons sake, a recently completed study of municipal
sewage sludge agricultural utilization programs at 9 sites showed
a range of annual sludge application rates of 15.3-146 dry Tons/ac
The median value was_ 50 dry Tons/ac (63).
VII-20
-------
The sodium adsorption ratio (SAR) i-s also an important con-
sideration in the application of a liquid waste to soils. Cal-
culation of SAR can be made by the following equation:
SAR
/Id
Z
Wastewaters with SAR values greater than 15 should be avoided
because of their detrimental effect on soil structure and ultimate
reduction in the- infiltration rate of the soils. Sodium adsorp-
tion ratio values from 5 to 15 can, over a period of years, lead
to loss of structure in soil horizons containing more than 10 or
2.0 percent clay (loam or finer texture). Lower values are gener-
ally satisfactory, although long term declines in infiltration
and percolation capacities have been observed in moderately fine-
textured soils when irrigated with water having SAR ratios as Tow
as 3 ( 54} .
The. SAR value calculated for mostos is 1.74, suggesting that
long range application to soils will not result in degradation of
its structure. Further, and of more practical significance, is
that both Bacardi in Tultitlan, Mexico and Serralles in Mercedita,
Puerto Rico, have practiced a "combined" form of agricultural ;
utilization/land farming over a time frame considerably longer
than calculations show to be feasible, yet neither has reported
any deleterious effects to soils or crops.
/
i*
Hydraulic Loading--
The ability of a soil to assimilate a specific quantity of
waste is a function of many variables, including:
VII-21
-------
Application rates "
Waste characteristics
Soil type
t Location of water table
Rainfall
Evaporation-
Twelve (12) ac-in is a suggested annual application rate-.
The exact. mostos application rate suitable to each location would
have to be determined in a pilot test program. Application rates
on St. Croix, for example, may be several times greater than in
Puerto Rico because of the lower precipitation and higher evapora-
tion rates. Mostos could be applied on a regular basis (1 ac-in/mo)
or an equivalent schedule to cope with weather patterns. Addi-
tionally, storage facilities would have to be constructed for reten-
tion during periods of inclement weather.
The soils of the coastal plains in and around San Juan and
Arecibo range from poorly to well-drained. In some areas, perched
ground waters may reach the surface during the wettest portions of
the year. On slightly higher ground, improved drainage characteris--
tics would be expected. Evaporation exceeds rainfall throughout
these coastal areas by about 22 in.
St. Croix, on the other hand, is semi-arid and has suffered
under drought conditions for the past few years. The soils sur- .-.
fm
rounding the VIRIL plant are mainly composed of the Diamond ser-
<+
ies (soil which is nearly level to moderately sloping, well drained
and shallow over semi-consolidated limestone). The subsoil is
VII-
-------
about 15 percent limestone fragments. Hard limestone is at a
depth of about 14 in. Evaporation exceeds rainfall by about 40
to 45 in (.84) .
The last four factors are site specific and while detailed
examination of each is not within the scope of this preliminary
study, certain general comments are applicable.
Land Requirements--
Table VII-6 presents the relationship between minimum land
requirements and application rates for each of the three dis-
ti11eries.
Land requirements are also directly affected by the solids
content of mostos, i.e., as the solids content increases, the land
requirement decreases. No attempt was made to correlate these,
however, since the only practical method of increasing solids
is by concentration through evaporation, implying a twofold treat-
ment scheme.
In reality, the estimates shown in Table VI.I-6 must be increased
somewhat to cover contingencies such as extended periods of soil
saturation, equipment storage, access roads, protective runoff
ditches, etc.... These considerations will be elaborated in the
Conceptual Design section of this report.
Bacardi--From the data presented in Table 6 and expanded to'.
'*>
account for the abov-e mentioned factors, Bacardi would require
/
220 to 317 ac to implement a land farming program. The lower
figure would correspond to an annual application rate of 18 ac-in;
the higher acreage, to 12 ac-in.
VII-23
-------
TABLE VII-6. MINIMUM ESTIMATED LAND REQUIREMENTS FOR
BACARDI, PUERTO RICO DISTILLERS AND VIRIL AS A
FUNCTION OF APPLICATION RATES
Land
Plant
Bacardi
Puerto Rico
Distil 1 ers
VIRIL
Required for Annual
9 ac-in
(acres)
370
165
80
*
Application Rates of :
12 ac-in
275
125
60
18 ac-in
18.5
83
40
*Based on assumption that heavy metal content and volume produced
remain constant.
Puerto Rico Distil1ers--Similarly, Puerto Rico Distillers
would require 121 to 155 ac for the same application rate (12 ac-
in/yr).
VIRIL--Land requirements for VIRIL are 52-80 ac.
CONCEPTUAL DESIGN OF A LAND FARMING PROGRAM
General
Land farming is a viable and economically feasible option for
the disposal of many and varied waste forms. The following analysis
presents a conceptual design of a landfarming program-for each of
the three distillers. Specific site information, operating data,
economics, etc., are_ unknown. Therefore, certain engineering
assumptions have b-een made and are documented throughout.
State agencies generally prefer land farming sites to have a
maximum slope of five percent. The minimum desirable depth to
VII-24
-------
groundwater is typically considered to be 15 ft, although this cri-
terion may be modified depending on the quality and uses of ground-
water and the types of overlying soils. It is also preferred that
the soils be. deep - greater than 4- ft - and moderately oermeable.
Waste and site characteristics assumed for the conceptual design
are summarized in Table VII-7. The developmental design is based on
land farming of mostos.
To provide cost comparisons for differing quantities of mos-
tos disposal, the conceptual design considers three annual dis-
posal rates: 33,100 dry Tons (Bacardi); 14,700 dry Tons (Puerto
Rico Distillers); and 7,350 dry Tons (VIRIL). While these figures
are only estimates, the conceptual design and resultant costs
clearly show the economies of scale for tile three rates con-
sidered.
A sketch of the hypothetical land farming site is shown in
Figure VII-1, The topography gf the site i.s, such that slopes range
from one to three percent, with an average's1 ope of approximately
two percent. The site is part of an open drainage system, which
is typical of most humid regions. As a result, the movement of
sediments and soluble materials from the site to neighboring water
courses is possible. However, the site can be modified to control
runoff. '£.
f*
Soils at the Jrypothetical site are medium textured, such as
silt loam, with a clay content ranging from 10 to 25 percent and
varying with depth. The soils are moderately well drained and
have an available moisture holding capacity of approximataly 15
VII-25
-------
TABLE VII-7. BASIS FOR CONCEPTUAL DESIGN
Waste characteristics:
t Chemical and physical characteristics as shown in Table 1
Annual generation rates as shown in Table 2.
Site characteristics:
blppe from 1 to 3 percent, averaging 2 percent
Silty loam soil at least 4 ft deep
Soils are moderately well drained, with a moisture holding capacity
of approximately 15 percent
Average precipitation of approximately 60 in/yr in Puerto Rico,
and 40 in/yr in St. Croix
< At least 15 ft to groundwater.
ii
i^j Application rate:
en
12 ac-in (120 dry tons/ac) per year.
Si te moni toring:
Six water samples taken quarterly
- Two surface water samples
- Four groundwater samples
Two soil samples taken quarterly
- One surface 0- to 12-in depth or within plow layer
- One subsurface 12- to 24-in depth
Ten water quality parameters measured in each sample.
-------
ro
Figure;. V.I I -1. Artist's conception of land cultivation site
-------
percent. The site is assumed to have been cleared and used for
pasture or low level agriculture for several years.
Land area required depends on the mostos application rate
which normally must be determined on a case-by-case basis. The
assumed mostos characteristics are such that an annual applica-
tion rate of 120 dry Tons/ac is appropriate for a useful site life
of 25 years. At this application rate, the productivity of the
site for agricultural crops after 25 years of land farming may not
be irreversibly impaired. It is preferable, however, that the
site be used solely for disposal purposes; no crop is grown dur-
ing or after completion of disposal operations to prevent intro-
duction of toxic substances into the food chain. Further, from
what has been presented earlier, the actual site life may be con-
siderably larger than that calculated as evidenced by the Serrales
and Bacardi, Mexico land disposal operations.
A surface and groundwater monitoring program is a necessary
part of any land farming design. To adequately monitor surface
water near or adjacent to a site, two samples should be taken
at quarterly intervals. One sample point should be located
upstream from the site and the other downstream.
The number of wells required to adequately monitor ground-
water at each site depends on the complexity of the subsurface -'
i
hydrology. For each hypothetical site, four wells are specified.
Two wells are locfated at the upstream boundary of the sits to
establish background water quality. Two wells are located down-
stream from each site to establish the distribution and impact
VII-28
-------
of localized contamination (if any) on water quality in the aqui-
fer. More wells would be desirable, but cost considerations will
usually indicate that a plan be developed to ensure most effec-
tive coverage with the least number of wells.
The suggested soil monitoring program entails taking 10
soil samples each at the Q to 12 in and 12 to 24 in depths
prior to the first mostos application and at quarterly intervals
thereafter. The samples at each depth are composited, processed,
and anlyzed. Data obtained for the soils which have received
mosto can be compared with the results for the controls (samples
obtained prior to first mosto application). This comparison will
indicate any accumulation and the extent of vertical migration of
most constituents (e.g., heavy metals) beyond the plow layer. It
is recommended that the following parameters be monitored'-
r Total dissolved solids (TDS) or soluble salts (EC)
TOC
pH
Sulfates
Chlorides
t Nitrate-nitrogen
Iron
Copper
Potassium
/-
Cadmium
Other parameters of concern such as selenium, boron, and
toxic organics should be included in the water and soil mor.itor-
VII-29
-------
ing programs if these constituents are present in significant
concentrations in the incoming mostos.
All analytical determinations can be performed by a contract
laboratory.
Quarterly water sampling should be adequate for monitoring
site performance. Such a sampling frequency has been typically
employed by numerous land farming site operators and state regu-
latory agencies.
Site Design
For the purposes of the following discussion, it is assumed
that the site selection process has been completed. Since each
site's topography is gently sloping and relatively flat, substan-
tial grading is not required to facilitate land farming operations.
The site has been used previously for pasture or low level agri-
culture, so substantial clearing is unnecessary. Uncontrolled
sheet runoff from the site's surface during periods of intensive
rainfall could result in contamination of adjacent surface waters.
For this reason, construction of a system of berms and runoff
collection ditches around the site perimeter prior to initiating
land farming activities fs specified. This surface drainage sys-
tem will eliminate standing water in the cultivated area, thereby
ensuring that aerobic processes necessary for mosto decomposition
are maintained. '
Appropriate liming of the soil or neutralization of mosto may
be necessary to maintain a soil pH between 6.5 and 7.5. Addition-
ally, application of nitrogen fertilizer may be needed if the
waste is highly carbonaceous (carbon to nitrogen ratio <30) to
VII-30
-------
enhance decomposition. These parameters will have to be inves-
tigated further in a pilot plant program.
An access control fence is placed outside the disposal area's
ditch/berm and within 100'ft of the site perimeter. A 30 ft gate
is located where the access road enters the site.
The site includes a paved access road to an office/equipment
storage building. The building is prefabricated aluminum on a
cement slab, with plumbing and utility connections. A 10 x 20
ft office extension are-a is furnished for the a.ttandant and includes
restroom facilities. The remainder of this structure' houses the
cultivation equipment and is closed to the weather on three sides.
Mostos Application Procedures
.Selection of an application system depends upon several fac-
tors including facilities at the distillery, transportation costs,
required flexibility, capacity and regulatory agency requirements.
In most instances, regulatory agencies require that wastes be
covered or mixed with soil soon after application to prevent
aesthetic problems and contamination of runoff waters.
Subsurface injection has proven successful in many situations.
These systems are normally used with liquid's in the range of 1 to
10 percent solids. Although subsurface injectors particularly
the continuously operating types, can apply liquid wastes at rela-
J?,>
tively high rates, factors such as transportation methods, costs
^»
and storage faci1fties for inclement weather require serious con-
sideration.
Selection of a truck- or tractor-mounted injector unit depends,
to a large extent, upon the goals established for the land farming
VII-31
-------
program, continuity of and access to the application site, and
the quantity of waste that must be applied in a given time period.
A truck-mounted unit offers the primary advantage of increased
flexibility. That is, since waste is supplied from a mobile tank,
the unit is essentially self-contained. It can be used both to
transport and apply waste and can easily be used to service several
sites. Typical productivity of truck-mounted units ranges from
30,000 to 80,000 gal/day and approximately 1,000 gal/acre per pass.
Primary advantages of the tractor-mounted unit are high pro-
ductivity (200,000 to 300,00 gal/day) and high application rates
(25,000 to 50,000 gal/acre per pass). Track-type tractors can
operate in more adverse weather and/or soil conditions than wheel
tractors or truck-mounted units. Tractor-mounted units, however,
require more sophisticated controls and waste handling facilities
to achieve maximum operating efficiency. Waste is supplied to the
tractor-mounted injector through a long flexible hose, and a fixed
liquid distribution system may also be required on the site.
Subsurface injectors normally use agricultural tillage tools
(chisels or sweeps) to open cavities in the soil, allow liquid
waste to flow into the cavities, and then cause the soil to bridge
or cover the openings. Sweeps generally provide better covering
at shallower injection depths and provide better mixing of the
waste and soil. Good mixing of the waste and soil at shallow
depths enhances evaporative drying, accelerates assimilation of
the waste into the soil environment and practically eliminates
odors, insects, and runoff problems. Further, in visual appear-
VII-32
-------
ance, the operation seems to be a normal agricultural practice.
Careful landscaping or screening of storage and other handling
facilities aids in eliminating the appearance of anything aes-
thetically offensive.
The following detailed analysis presents two scenarios: the
first utilizes a tractor-type application vehicle fed from on
site (land farm) holding tanks. Mosto is transported to the site
via tanker trucks. The second utilizes the same application
vehicle but on site mostos is supplied via a buried pipeline.
Via Tanker Truck--
Mostos is transported between the distillery and land farm-
fng site via 6,000 gal tanker trucks. A 10-day storage tank is
provided at each distillery to provide a buffer.during periods
of inclement wea.ther. A further 2-day capacity storage tank is
provided at each land disposal site to supply the application
vehicles. Application of mostos (assumed total solids of 3.35
percent) is accomplished using a subsurface injector mounted on
a crawler tractor, the discharge from which is rated at 800 gpm.
Application units are fed from the storage'tank through a dis-
tribution system consisting of 10-in buried cast iron pipe ter-
minating in a hose connect. The hose is heavy-duty flex type, -,.
" »
typically 600-700 ft in length and 4-in diameter
Table VII-8 s'hows the cultivation areas, quantities of mosto,
and storage capabilities required for each of the three sites
at the suggested 12 ac-in application rate. It is assumed that
the increase in the areas shown (total) will be required for
VII-33
-------
the access road, perimeter berm, buildings, runoff control facili-
ties, and a 100-ft buffer on all sides. Figure VII-2 presents a
representative profile of the site, indicating several of the
key features previously discussed.
Mostos is applied on a daily basis except in extended periods
of heavy precipitation during which the soil will not support the
weight of the crawler tractor. It is assumed that soil applica-
tions can be made an average of 210 days/year and that application
will be restricted to a one shift/day operation.
TABLE VII-8. CONCEPTUAL DESIGN WASTE AND SITE PARAMETERS
Parameter?
Estimated annual volume (gal)
Gallons per operating field day
Bacardi
90,000,000
423,000
Puerto Rico
Distillers
40,000,000
190,000
VIRIL
20,000,000
95,000
(assume 210 working days/yr)
Estimated annual total solids
disposed.(? 3.35% solids) (Tons)
33,066
H,S96
Land area required
(12 ac-in/yr):
.acres
7,343
For mostos disposal
* Total (including area for access
road, building, ditches, buffer
zone, etc.)
Storage tank volume at distillery
with 10- day storage capacity
(mill ion gal )
Storage tank volume at land
farm sita with two-day storage
capacity (million gal].
275 125 SO
317 155 80
4.3 1.9 1.0
0.86 0.38 0.19
VII-34
-------
FENCE
TOO F
BUFFER
ZONE
EQUIPMENT STORAGE
BUILDING AND OFFICE
FENCE
LAND CULTIVATION FIELD
RUNOFF
CONTROL
DITCH fc
BERM
MONITORING WELL
RUNOFF
CONTROL
DITCH t
BERM
Figure VII-2. Representative site profile
-------
Truck transport of mostos in 6,000 gal tractors will operate
on an extended (8 or more hours/day) schedule, approximately 210
days/year, (more for Bacardi) to maintain feedstock to the field.
Times to load at the distillery, travel to the site, unload, and
return are estimated to be (assuming site is five miles from
distillery) 10, 15, 10, and 15 minutes, respectively. Further,
it is estimated that this operation will be 70 percent efficient,
therefore requiring a total cycle time per tanker of 71 minutes.
Based on the above (presented in Table VII-9). Bacardi will
require nine tractor trucks, Puerto Rico Distillers four, and
VIRIL, two. Each, will require one additional cab for preventive
maintenance and nonscheduled repairs. Similarly, at a rated
capacity of 800 gpm, Bacardi will need two crawler-tractor applica
tion vehicles to handle their daily disposal volume; both Puerto
Rico Distillers and VIRIL will require only one each. Labor
requirements basically will follow the manning of equipment pieces
as shown in Table VII-10.
VII-36
-------
TABLE VII-9. CONCEPTUAL DESIGN VEHICLE REQUIREMENTS FOR LAND FARMING
UTILIZING TANKER TRUCKS FOR iMOSTOS TRANSPORTATION
Factors
Puerto Rico
Bacardi Distillers
Gal/land farm operating day (gal)
Tanker loads/day land farm @ 6,000 gal
per tank
Total cycle time for one tanker load
@ 70* efficiency (rain)
Number of tank trucks required per day
assuming one shift - 10 hr/day
Actual number of tanker vehicles
required:
428,000
72
71
8.4
190,000
32
71
3.7
TABLE VII-10.. LABOR. REQUIREMENTS FOR LAND
FARMING UTILIZING TANKER TRUCKS FOR
MOSTOS TRANSPORTATION
YIRIL
95,000
16
71
1.9
Tankers
Cabs
Number of tractor-crawler field
applicators required at a.
capacity of 800 gal/min
9
10
2
4
5
1
-2
3
1
Tanker Truck Drivers
Crawler-tractor Drivers
Full Time Mechanics
Foreman
Relief Personnel >
Total
Bacardi
9
2
2
1
1
15
Puerto Rico
Disti 1 lers
4
1
1
^
-
6
V I R I L
2
1
-
3
*Assume that employee assigned to crawler-tractor can
substitute as a part-time mechanic.
VII-37
-------
Via Pipeline--
This scenario is similar in all respects to that presented
in the previous section with the following exception: mostos
is supplied to the land farm site via an 8 to 10 in diameter
b.uried pipeline.
Mostos storage is assumed to remain the same - i.e., a ten-
day capacity tank at the distillery and a two day capacity tank
at the land farm site. Labor requirements, revised from Table
VII-10 accordingly, are presented in Table VII-11.
TABLE VII-11. LABOR REQUIREMENTS FOR LAND
FARMING UTILIZING A PIPELINE
FOR MOSTOS TRANSPORTATION
Crawler-tractor Drivers
Full Time Mechanics
Total
Estimated Costs for Concep
Bacardi
2
3
. 5
tual Land Farming
Puerto Rico
Distillers
1
2
3
Site
VIRIL
1
1
2
Both capital and operating and maintenance costs are estima-
ted.
Capital costs associated with land farming for both scenarios
are presented in Tab-les VII-12 and VII-13. The recovery period
f»»
("useful life" or "payback" period) varies according to the type
of investment. All capital costs have been annualized based on an
11 per-cent interest rate (on internal rate of return) over their
VII-38
-------
TABLE VII-12. CAPITAL COSTS FOR THREE CONCEPTUAL LAND FARMING SITES
UTILIZING TANKER TRUCKS FOR MOSTOS TRANSPORTATION*
Cost Element
Debt service for
purchase of land @
Recovery
Period
(yr)
Bacardi
Total Annual
523,000
(317 ac)
Puerto Rico
Distillers
Total Annual
255,800
(155 ac)
VIRIL
Total Annual
132,000
(80 ac)
11% yr; ba'secl on
original price of
$15,000/ac
Site preparation: 20 95,100 11,900 46,500
Leveling, some
clearing, survey-
ing, design @
$300/ac
Site work: 20 349,000 43,000 171,000
Fencing, build-
Ings, roads, run-
off prevention
ditch & $l,100/ac
Monitoring wells: 20 4,000 500 4,000
4 to 7 1n wells,
60 ft deep, In-
cluding pumps and
casing & $1,000/
well
Storage tanks &
$.30/gal.
At the distillery 20 1,290,000 162,000 570,000
At land farm 258,000 32,400 114,000
5,840
21,400
500
71,600
14,300
24,000
88,000
4,000
300,000
57.000
3,010
11,000
500
37,700
7,160
-------
TABLE VII-12 (continued)
o
Recovery
Period
(yr)
10
Cost Element
Field distribution
system:
From farm storage
to field stations;
includes pumps,
risers, underground
pipe @ $l,000/ac
Electrical equipment: 20
Includes installa-
tion of power at site,
transformers, con-
trols & $600/ac
Property tax and
insurance & 1.5% of
in-place capital
investments
Injection equipment;
Includes crawler-
tractor, hoses,
reels, freight
Communication
equipment
Subtotal
10
Bacardi
Total
5,000
Annual
317,000 53,800
190,200 23,900
35,200
240,000 40,000
1,350
Puerto Rico
Distillers
Total
155,000
93,000
2,748,300 928,650
120,000
5,000
1,278,500
Annual
26,300
11,700
16,605
20,400
1,350
445,795
VIRIL
Total
80,000
48,000
120,000
5,000
726,000
Annual
13,600
6,030
8,650
20,400
1,350
241,400
-------
TABLE VI1-12 (continued)
Cost Element
Tanker trucks &
Recovery
Period
(yr)
10
Bacardi
Total Annual
580,000 98,500
Puerto Rico
Distillers
Total Annual
280,000 47,500
VIRIL
Total Annual
180,000 30,600
$60,000/unit
($40,000-cab;
$20,000-tanker)
Total Capital
Investment
Annual Capital
Costs
Annual Capital
Costs/Dry T
Annual Capital
Costs/1,000 gal
Mostos
3,328,300
1,027,150
1,558,500
31.06
11.41
423,295
33.57
12.33
906,000
272,000
37.03
13.60
*"Total" amounts are initial costs, "annual" amounts are for first 12 months of operation. All costs
are in 1978 dollars.
Annualized capital recovery (equal annual payments) at 11% interest rate.
-------
I
-F".
ro
TABLE VII-13. CAPITAL COSTS FOR THREE CONCEPTUAL LAND FARMING SITES
UTILIZING A PIPELINE FOR MOSTOS TRANSPORTATION*
Recovery
Cost Element* Period (yr)
Subtotal from Table 12
Pipeline: 5 ml of 6" dia. 20
pipe, including acces-
sories e $ia/ft
Pump stations: 2/slte 10
@H 50.000 ea
Debt service for right 20
of way lease & $15.000/ac
(lit Interest)
Property taxes and
insurance $ 1.5% of 1n-
place capital invest-
ments
Total capital Investment
Annual capital costs
Annual capital costs/
dry T
Annual capital costs/
1,000 yal mostos
Oacardi
Total Annual
$2,74B.300 $ 920.650
343.200 43.100
300 .000 51 .000
16,500
9.650
$3.391.500
$1,040.900
$ 31.72
$ 11.65
Puerto Hi co Distillers
Total Annual
$1.270.500 $445.795
343.200 43.100
300,000 51.000
16.500
9.650
$1.921.700
$566.045
$ 30.52
$ 14.15
VIIIIL
Total Annual
$ 726,000 $241,400
343.200 43.100
200.000 51.000
16.500
9.650
$1,369.200
$361 .650
$ 49.23
$ 10.00
'"Total" amounts are initial costs, "annual" amounts
are for first 12 mo of operation. All costs are
In 197U dollars.
t/Wmualized capita).recovery (equal annual payments)
at 11* interest rate.
-------
recovery period. Major capital cost categories are land, site
preparation and construction, and equipment.
The estimated unit cost for land is $15,QOO/ac. However,
land is considered an appreciating asset and as such is not con-
sidered in the same light as other investments. Rather it is
shown as an annual cost in the form of debt service.
Tables VII-H and VII-15 respectively present annual capi-
tal and operating costs for both scenarios. From the data shown,
lower total annual costs are. shown for pipeline transport for
Bacardi and Puerto Rico Distillers. From the scenarios presented,
the less expensive option to VIRIL would appear to be tanker trans-
port. This would change, however, if VIRIL would be able to
utilize the adjacent land surrounding the plant (.the scenario is
for a disposal site five miles from the plant.
Table 711-16 presents a summary of Tables VII-12, VII-T3,
VII-H and VII-15.
In considering the foregoing analysis, however, several
qualitative factors should be included when comparing the scenarios
These are:
Social and aesthetic cost are intangible but should be
considered. For example, the logistics of Bacardi hauling
70 or more daily loads (6,000 gals each) of mostos on -:;
-X
reasonably tuisy roads would suggest considerable traffic"
congestion-^ noise, and possible annoyance to roadside
vi 11 agers.
Further, no costs have been included for highway and road
repairs. These are difficult, at best, to calculate, but
VII-43
-------
I
I.
TABLE VI1-14. ANNUAL OPERATION AND MAINTENANCE COSTS FOR THREE
LAND FARMING SITES UTILIZING TANKER TRUCKS FOR MOSTOS TRANSPORTATION
Cost Element
Labor @ $10/hr, Including fringe
benefits
Operation and maintenance of vehicles
@ 20% of capi'tal costs; includes fuel
and replacement parts
Monitoring (32 samples x 10 parameters
ea @ $20/parameter plus other testing)
Annual operating costs
Annual capital costs
Subtotal
Administrative costs @ 15%
Total annual costs
Annual costs/dry T
Annual costs/1,000 gal mostos
Bacardi
$ 312,000
164,000
10,000
486,000
1,027,150
1,513,150
226,970
1,740,120
52.63
19.33
Puerto Rico
Distillers
$ 124,800
80,000
10,000
214,800
493,295
708,095
106,214
814,309
55.41
20.36
VIRIL
$ 62,400
60,000
10,000
132,400
272,000
404,400
60,660
465,060
63.31
23.25
-------
TABLE VI.I-15. ANNUAL OPERATION AND MAINTENANCE COSTS FOR
THREE LAND FARMING SITES UTILIZING A
PIPELINE FOR MOSTOS TRANSPORTATION
Cost Element
Labor @ $10/hr, including fringe
benefits
Maintenance:
t of pipeline @ 2% of
capital investment
of pump stations @ 2># each
of capital investment
Operation and maintenance:
t of crawler- tractor vehicles
@ 20% of capital investment,
including exchange parts
and fuel
Power and supplies for operation of
pump stations @ $15,QOO/yr
Monitoring (see Table 14)
Annual operating costs
Annual capital costs
Subtotal
Administrative costs @ 15%
Total annual costs
Annual cost/dry T
Annual cost/1 ,000 gal mostos
Bacardi
$ 104,000
6,860
7,500
40,000
15,000
10,000
183,360
1,048,900
1,232,260
184,839
$1,417,099
$ 42.86
$ 15.75
Puerto Rico
Distillers
$ 62,400
6,360
7,500
20,000
15,000
10,000
121,760
566,045
687,805
103,170
$790,975
$ 53.82
$ 19.77
VIRIL
$ 41 ,600
6,860
. 7,500
20,000
15,000
10,000
100,960
361 ,650
462,610
69,392
$532,002
$ 72.42
$ 26.60.
VI1-45
-------
TABLE VI1-16. SUMMARY OF ANNUAL CAPITAL AND
OPERATING TOTAL COSTS FOR TANKER TRUCK VS.
PIPELINE MOSTOS TRANSPORTATION
Cost Element
Tanker truck transportation:
Annual capital costs/dry T
Annual capital costs/1,000 gal
mostos
Annual operating costs/dry T
Annual operating, costs/1,000 gal
mostos
Bacardi
$31.06
11.41
52.63
19.33
Puerto Rico
Distillers
$33.57
12.33
55.41
20.36
VIRIL
$37.03
13.60
63.31
23.25
Pipeline transportation:
Annual capital costs/dry T
Annual capital costs/1 ,000 gal
mostos
Annual operating costs/dry T
Annual operating costs/1,000 gal
mostos
31.72
11.65
42.86
15.75
38.52
14.15
53.82
19.77
49.23
18.08
72.42
26.60
VII-46
-------
could become significant, especially so for roads designed
for normal passenger car traffic - not fully loaded tanker
trucks.
« Annual capital costs for both scenarios can be significantly
reduced if the cost of land is less than shown. Addition-
ally,, long-term leases may be negotiated at a fee much
less than that used for debt service.
No costs are presented, for appreciation of land values.
If purchased, this land would be expected to increase
in value, thus increasing the net worth of the respective
purchasing companies.
* All estimated costs for the conceptual designs are based
on 1978 dollars. It is to be expected that those costs
will rise in the future, du.e to inflation, but no cost
projections were made because of uncertainty in the infla-.
tion rate.
The annual operating costs of mostos pipeline transport
(Table Til-15) are 14 percent of -the to ta^l. annual costs. The sane
values for tanker truck transport show the annual operating costs
to be 47 percent (Table VII-14). The comparison is significant since
those operating dollars needed for annual support of a tanker
truck fleet,would be expected to increase with inflation. Total
'i
operating costs for pipeline transport are similarly subject
/
to inflation, but'.would be expected to increase less in overall
total dollars since they represent a smaller overall percentage
of the total annual costs. The relative cost of pipeline trans-
port, therefore, would be expected to become more economically
-------
favorable in the ensuing years, assuming that the inflation rate
continues to increase.
As expected, the total unit costs shown indicate that larger
land farming sites are less costly to operate on a unit cost
basis than are smaller sites. For comparisons the range of unit
costs on a dry weight basis for either scenario are within the
range of municipal sewage sludge disposal costs typically repor-
ted.
Finally, Table VII-17 presents the effect of each scenario in
terms of annual proof gallon production (1977 figures). It was
assumed that production figures were 17.4, 5 .2 , and. '3..'4 mi 11 ion
proof gallons respectively for Bacardi, Puerto Rico Distillers,
and- VIRIL.
TABLE VII-17. ANNUAL LAND FARMING COSTS
OF MOSTOS AS A FUNCTION OF
RUM PRODUCTION
Scenario A - Transport
via Tanker Truck
Scenario B - Transport
via pipe!ine
Puerto Rico
Bacardi Distillers VIRIL
$/proof gal Ion
0.1000 0.1566. 0.1368
0.0814 0.1522 0.1565
VII-43
-------
IDENTIFICATION OF RESEARCH NEEDS
The preceeding sections have outlined the basic fundamentals
of a land disposal program and have carried these through in the
form of a conceptual land farm design. Throughout the discussions,
certain key parameters were identified, which, if quantified, would
facilitate a- more detailed and precise feasibility evaluation. A
pilot study would, therefore, seem a logical first step to obtain
the required data. The. following program describes the equip-
ment, site preparation, mos.tos applications, and monitoring recom-
mended to determine the acceptability of the landfarming prac-
tices described in the Conceptual Design section.
A prerequisite, however, to this en.tire program is a thorough
exploration with local regulatory agencies for necessary permit-
ting and acceptance of the concept.
Equipment Required
For the conduct of this small pilot program,it is recommended
that the necessary equipment.be rented, either with or without an
operator. For best results, the same type of equipment which will
be used for the full-scale land farm should be used for the pilot
test program. This, however, is not practical in all cases,
expecially so with reference to subsurface injection equipment.
>
In lieu of not having these specialized pieces, it is recommended;
r'
that conventional surface applications followed by disking be sub-
/
stituted. The results are generally comparable. It is anticipa-
ted that the following equipment will be required:
VII-49
-------
t Farm tank trailer (1,500.- 2,000 gal capacity) equipped
with exit valve and bar or fan spreader
Tractor
Roto-tiller or disk (.3-tier, 10 ft wide, 18 in diameter).
The program should be implemented on a small plot of land (10 to
15 ac) and should be allowed to run for 6 to 9 mo, or an equivalent
time frame to cover the seasonal variations. The location of the
site is preferably near either the Bacardi or Puerto Rico Distillers
facility. It may be assumed with a certain degree of confidence
that application rates and other parameters determined in Puerto
Rico will be applicable to a similar operation on St. Croix. The
opposite, however, may not be the case considering the current
drought conditions of St. Croix.
The test and control plots should be constructed so that they
are of equal size and are located on identical soils. A berm
should be constructed around the perimeter of the test area and
between the test plots to contain runoff and mosto applied within
the fi.elds.
Mostos Application
In the pilot program, mostos should be applied at rates equi-
valent to 6, 9, 12, 15, and 18 ac-in/yr to each of 5-two ac plots.
A sixth 2-ac plot would serve as a control (no mostos would be .^
added to this plot 'for the duration of the testing period). Appli-
cations can be made by filling the tank (known volume) with mostos,
driving to the particular plot designated for the specific appli-
cation rate, and subsequently discharging the \oad at ?. previously
VII-50
-------
calculated speed equivalent to the prescribed application rate.
Depending on the weather, the mostos 'will be allowed to dry for a
few days (or more if it rains), and then mixed with the surface
soils to a depth of 6 to 8 in, with a roto-tiller or tractor-
pulled disk. Disking should be accomplished after each applica-
tion and may have to be increased if odor problems develop.
Mostos should not be applied when the soil is excessively
wet, since the equipment may get stuck in the mud and odors may
possibly develop from casual ponding.
To facilitate evaluation of the pilot test .program, ' precise
records need to be maintain-ed on the time, quantity and percent
solids at which each load of mostos is applied to the plots. In'
addition, cultivation records need to be kept.
Monitoring
Soil characteristics and hydrogeologic data for the sita
should be carefully investigated. Of particular concern are such
factors as permeability, soil pH, and distance to groundwater.
Major environmental concerns typically, however, are soil accumu-
lations of heavy metals and the potential for surface water con-
tamination .
In the test program, soil samples from 0 to 1 ft, 1 to 2
ft, and 2 to 3 ft depths should be collected prior to the first 4
mostos application.^ Subsequent sampling should be conducted at
i"
three-month intervals. These samples are to be analyzed for heavy
metals (e.g., Cd, Pb, Cd, In, etc.), organic matter, total Kjeldahl
N, pH. nitra.te-N, chlorides, and sulfates. If these constituents
VII-51
-------
show significant increases in concentration with waste application,
the extent to which they build up in soil and potential for sur-
face water contamination can be assessed.
Likewise, runoff from the plots should be collected periodi-
cally and analyzed for the above-mentioned constituents.
Odor problems, if any, are to be noted during daily inspection
of the plot, particularly after spreading of the mostos and on
wet days.
Of special concern is the effect of mostos on soil pH. As
mentioned earlier, mosto is very acidic (pH = 4.4) and may require
neutralization prior to application. Alternately, liming of the
soils may be required.
VII-52
-------
VIII. . EVAPORATION
Evaporation of mostos, when viewed as a treatment technology,
must be examined as a function of two considerations:
Available technology
». Disposal options for secondary products (concentrates
and side waste streams).
The technology for evaporation of cane molasses stillage
has been successfully, demonstrated (57, 58, 59, 60). Typically,
the end result of such treatment is the production of a secondary
product known as condensed molasses solubles (CMS) with a solids
content in excess of 50 percent.
European and United States distillers have demonstrated
that CMS can be successfully marketed. Puerto Rico and the Virgin
Islands, however, are islands with limited potential for domestic
CMS consumption (see "Potential for CMS Use in Puerto Rico and
the(U.S. Virgin Islands" for a detailed discussion). An in-depth
analysis of evaporation as a disposal option, therefore, will be,.
directed toward discussions of both the technologies available to
^
produce secondary products and the alternatives available to dis-;i
pose of these (fin
-------
hearings in the matter of NPOES permits for Puerto Rico Dis-
tillers, Schenley, and VIRIL ( 15» 24, 34, 87 and 57-60). Similar
experience has also been reported by Bacardi ( 4). This testimony
is assumed to be accurate and reliable. Included in the testimony
are some cost estimates which are typically subjective judgements
of the party offering testimony. The reader should keep in mind
that these estimates are provided for general information without
substantiation. Additional technological input to the various
processes employed has been supplied by equipment manufacturers.
Theory, Design, and Operation
Two basic evaporation principles have successfully been
utilized by the distillery industry to treat liquid "waste streams -
multiple effect and mechanical recompression. Within each cate-
gory, various evaporator types are available to handle evapora-
tion duties with maximum efficiency and economy. The most com-
mon types currently in use by the distillers are either plate
(falling or rising film) or tubular evaporators (falling film).
The following section will briefly discuss each of the above and
their interrelationships.
The principle of multi-effect (ME) operations allows several
single evaporation units (known as effects) to be connected in !
series. This system involves the multiple reuse of the heat con-
tent of steam. By ,this method, the energy contained in a single
pound of steam may be used to evaporate anywhere from two to eight
pounds of water, depending on the number of effects and the aux-
iliary equipment employed. Steam from a boiler enters the heating
VIII-2
-------
jacket in the first of a series of interconnected evaporators.
This steam boils the mostos in the first evaporator and conden-
ses in the heating jacket after it has transferred its heat con-
tent to the mostos. This condensate is returned to the boiler.
The water vapor from the boiling mostos is piped to the heating
jacket in the second evaporator, where it boils the mostos in
that evaporator. This process is continued as many times as
the number of effects in the system (3).
Mechanical vapor recompression (MVR) entails increasing the
heat content of the water vapor from the evaporation process.
The recompression is generally applied to the vapor from the first
effect in the system, causing an increase in the heat content of
the vapor and.enabling it to evaporate more water per initial
unit of heat input (3). Some manufacturers produce MVR evapora-
tors only as single effect units (3).
Generally, the evaporators which would be considered for
application to mostos are designed on the basis of either thin
film or forced circulation principles (3, 28). Five basic eva-
porator designs employ the thin film principle (3):
The falling film tubular evaporator
The rising film tubular evaporator
The falling film plate evaporator ^
.-»
i The free flow falling film evaporator.
i»'
Thin film systems^are offered as single-or multiple-effect units,
with or without MVR.
V 111 - 3
-------
Forced circulation evaporators are designed for processing
liquids which are susceptible to scaling or crystallizing (74,75).
Two basic types are offered (3):
The forced circulation tubular evaporator
The forced circulation plate evaporator.
Forced circulation systems are offered as single-or multiple-
effect units, and may be combined with film evaporator pre-
concentrating effects. MVR can be considered with forced cir-
culation systems, but economic considerations usually do not
favorsuchadesign (28).
The energy efficiency of an evaporation system can be
increased through the use of additional effects and MVR, but
only at the cost of increased capital outlays from the. addition
of such equipment. MVR is theoretically the most energy-efficient
evaporation method (34). MVR provides its greatest energy cost
savings in applications where the cost of electricity is low and
the operating hours per year are relatively high.
Applicability to Mostos
Evaporation of molasses stillage is a common occurrence in
Europe, but, with few exceptions, it is applied to beet instead
of cane molasses stillage. Two exceptions are the Bols Distillery
in the Netherlands and the Bacardi rum facility in Malaga, Spain;.
Bols has a four-effect evaporator of the falling film type with
a separate finishi-ng unit. This unit has been in use for four
years. The Bacardi, Malaga, evaporator, designed by Vogelbusch,
is a multiple-effect evaporator with forced circulation for final
concentration. Multiple-effect evaporators are currently in u'se at
VIII-4
-------
the Sols Distillery in Scheidam, the Netherlands; Florida Dis-
tillers, Lake Alfred, Florida; and Bacardi, Malaga, Spain. Prior
to plant closure, Standard Brands in Peekskill, New York, also
utilized multiple effect evaporation.
The Sols Distillery is not specifically a rum distillery,
but it produces liqueurs from the distillation of cane molasses.
The differences between liqueurs and rum result from processing
and formulation stages after distillation. Therefore, the com-
position of the raw waste discharge from the Sols Distillery is
similar to that from a rum distillation plant. The equipment at
the distillery is made of type 316-1 stainless steel to permit
cleaning with both dilute hot' nitric acid, as well as with dilute
hot caustic soda. Normally, four to six hours of cleaning are
required out of a 24-hour operating period to remove organic and
inorganic scale deposits (59).
Although Bacardi ' s m-ul ti pi e-effect evaporator at Malaga,
Spain, has been.in actual operation for a limited period of time,
it has successfully demonstrated the evaporation of rum slops (58)
Many initial scaling and fouling problems, however, were encoun-
tered. During initial start-up, one of the stages was in a con-
tinuous down-phase for cleaning.
In the United States, evaporation has been applied to beet,;'
Jr
citrus, and cane mojasses wastes from yeast, pharmaceutical, and"
<*
rum manufacturing.-faci 1 i ties (58). Standard Brands Industries
(S8I) successfully operated an evaporation system at its yeast
producing facility in Peekskill, New York. The primary waste
VIII-5
-------
stream from the facility consisted of the spent molasses slops
remaining after the fermentation process in the production of
yeasts. The raw material for the yeast production consisted of
varying amounts of cane and beet molasses. Restrictions in the
supply of beet molasses were such that SBI on occasion used only
cane molasses as their raw material. Although SBI initially
experienced difficulties with the evaporator when only cane molas-
ses was used, they could evaporate cane molasses wastes by imple-
menting some specific improvements to their system designed to
reduce the scaling (58).
The Red Star Yeast Plant in Belle Chase, Louisiana, produces
yeast from Brazilian cane molasses. Red Star evaporates the waste
to CMS and the evaporator needs to be cleaned only once every two
to three weeks (53).
Pfizer, a pharmaceutical manufacturer in Groton, Connecticut,
operates a large and technologically sophisticated evaporative
system for a beet molasses waste stream and only has to clean the
evaporators once every two weeks using caustic soda and nitric
acid (58),
Process Performa-nce
The overall evaluation of evaporation technology is a func-
tion of several parameters, the more important of which are:
Effect of process in reducing pollutants
t Operating-'eff iciency of equipment
Energy consumption
VIII-6
-------
Effect of Process in Reducing Pollutants--
According to the testimony of Erik Krabbe, Region II EPA
staff consultant to the Facilities Technology Division (60). the
rum producers are required by EPA to remove approximately 94 per-
cent of their waste-load. BOOg. The Malaga, Spain, Bacardi rum
plant is capable of 98.7 percent BODg reduction (58). Some data
from Florida Distillers in Lake Alfred, Florida, consisting of
seven samples taken between May 17 and 19, 1976, showed a mean
30D5 reduction of 96;.2 percent (53). Unfortunately, the Sols
Distillery does not measure SOOg removal.
Puerto Rico Distillers (15) has noted that the evaporation
process performance at Sols and other distilleries is greatly
affected by whether or not the yeast and the fermenter bottoms
are removed from the mostos stream. .At Bols they must be removed
to prevent rapid clogging of the evaporator. Puerto Rico Dis-
tillers has also noted that these high 800 sidestreams cannot
be discharged untreated to the ocean without violating the 300^
limits of their NPOES perm'it. Therefore, these yeasts and fermen-
ter bottoms must either be discharged to a POTW (an option not
available to any of the distillers - see Chapter V , "Disposal
to a Public Owned Treatment .Works (POTW)"} or may require fur-
ther treatment. :'.
*"S
The production--of CMS results in large quantities of evapor-
/»
ator condensate re'portedly having a SOD value of 100 to 2,000
pom (32). Supplemental biological treatment of this wastestream
may, therefore, be required prior to discharge. Erik Krabbe,
VIII-7
-------
however, has stated that biological treatment of the evaporator
condensate is not needed to meet the effluent standards required
by EPA (60).
Operational Efficiency of Equipment--
From a practical standpoint, the ability to successfully
evaporate mostos depends upon the nature of the input waste,
the desired output, and the operating conditions to achieve this
output.
It is generally accepted that the evaporation of cane molas-
ses stillage results in more severe scaling than either beet or
citrus molasses. For example, the Standard Brands operation in
Peekskill, New.York, was basically designed to evaporate beet
molasses. Their six-stage multiple plate evaporator, when pressed
into duty on only cane molasses (a condition which occurred for
a two-week period when beet molasses was not available), showed
evidence of even more severe scaling and fouling than normal (88).
Standard Brands, in more recent operations, has switched to
falling-film tube type designs (88).
Similar experiences were reported by the Zuid Nederlandse
Spiritusfabriek at Delfzil, Netherlands (ZNSF) (15). This plant
normally uses beet molasses,but on several occasions - two months
in 1974 and one week i.n 1977 - the plant had to exclusively use .
cane molasses. On the first occasion, six to eight' hours of
,' j
cleaning were requ'ired after running the evaporator for 24 hours.
During 1977, they were able to operate 10 to 15 hours before
cleaning, a sitation which more closely approximates the Bols
cycle (15).
VIII-3
-------
To alleviate the fouling problems, Standard Brands was
forced to modify their equipment. Their original APV-designed
evaporator was not able to concentrate beyond 40-percent solids
without experiencing problems. To overcome the basic design limi-
tations, they installed a two-stage falling-film tubular evapora-
tor to achieve final concentration to 55 percent.
Regardless of evaporator type, however, certain cleaning pro-
cedures must be employed on a regular basis to alleviate the con-
tinual build-up or inorganic and organic deposits. Typical ly, both
nitric acid and caustic soda are employed for removal of these
deposits. In cases of more severe scaling, mechanical cleaning
has to be employed.
Another factor affecting efficiency is the desired solids
content of the final product. Whereas CMS derived from beet molas-
ses may be taken to Q'5-70° Srix, CMS from cane stillage has been
generally limited to 55-60 Srix. Above this level, fouling and
.scaling become more pronounced and viscosity increases to a point
where the final product becomes difficult to handle.
Energy Consumption--
This was discussed in the section entitled "Theory, Design,
and Operation." .. .
Pilot Installation
$
Bacardi has bui-lt a pilot-scale evaporation system at its
<
distillery in Cataho, Puerto Rico, with an operating capacity of
approximately 30,000 gal/day (87). Bacardi has said that if the
VIII-9
-------
pilot tests are successful and a method can be found for disposal
of CMS, they may build a.full-scale system (34). The pilot tests
involve experiments with:
Construction materials
t Pretraatment alternatives
Cleaning cycles
Temperature and other parameter optimization
Condensata treatment
CMS marketing (34).
Bacardi's pilot-scale evaporator, a two-body, three-stage
type, was designed by the Dedert Corporation. The first two stages
are falling film; the final is forced recirculation . The system
was designed to operate on the principle of mechanical vapor
recompression (MVR). Bacardi believes that this will be the
best-suited evaporator for treatment of its waste from the stand-
points of cost and energy savings (34).
In Puerto Rico electrical power is six times more expensive
than steam energy. Therefore, Bacardi has opted to use a steam
turbine power generator to drive the MVR system rather than elec-
tricity (34).
It was calculated that the Bacardi pilot MVR unit would
require about 2.2 KwH/1 ,000 Ib water removed, while the Bacardi,'
Malaga, multiple-effect evaporator currently requires about 112
KwH/1,000 Ib water removed (57).
Bacardi will be the first to evaluate its efficiency in a
rum distillery for the treatment of cane molasses.
VI11 -10
-------
The construction of the pilot evaporation unit was completed
in August 1977, but the contractors were unable at that time to
furnish the turbine for the MVR. The unit was consequently run
with steam until July, 19-78, at w&Tcft time the MVR mode became
operative.
During the break-in runs, there was some evidence that scale
build-up was reduced, permitting longer operation periods between
cleanings. In one- of the early runs, for instance, the unit
developed some encrustation .after 120 hr of operation, and cir-
culation with a caustic solution did a fair job of cleaning (87).
It should be noted, however, that the long lengths of the initial
runs were chosen to study the outer limits of operation and may
not represent the most practical production-to-cleaning cycle.
Furthermore, the ash content of the molasses used was only 10
percent, whereas the average ash content of the molasses used by
Bacardi is approximately 12 percent (87).
Certain mechanical problems were also experienced during
several of the test runs, but these were corrected between the
runs. It was suggested that molasses pretreatment might provide
a waste that would be easier to evaporate, .but as of October
1978 this step has yet to be implemented (87).
The evaluations of both the evaporator and the MVR unit are-'
forthcoming from the Bacardi Corporation, but are not anticipa-
ted until sometime" after December 1, 1978. Until such a report
is 'issued, a detailed analysis of the units' performance is not
pos si ble .
vrn-n
-------
CONDENSED MOLASSES SOLUBLES (CMS)
UTILIZATION AS AN ANIMAL FEED SUPPLEMENT
The term "CMS" refers to a number of different products,
all of which are made from the residue of a molasses fermenta-
tion process. Molasses fermentation is involved in the produc-
tion of yeast, Pharmaceuticals, citric acid, monosodium glutamate,
alcohol (for both industrial and beverage use), and various other
products. During the manufacture of these products, the fermen-
table sugars contained in the molasses are consumed by micro-
organisms., leaving a residue which could be described as sugar-
less molasses. This residue is converted into CMS by evaporation
until a concentration of 55 to 60 percent solids is attained (35).
The composition of CMS varies greatly depending upon the
type of molasses used and, to a lesser extent, the type of fer-
mentation process and end product involved. The single most
important determining factor in the composition of CMS is whether
the input material is cane or beet molasses. In addition, there
is considerable variation when cane molasses is used because
cane molasses varies greatly in composition, depending upon the
weather and soil conditions where the cane was cultivated, the
method by which the cane is harvested, and the efficiency of the
sugar-refining process, of which molasses is a by-product (35). '$
Beet-based CMS is much more desirable as a feed additive
than cane-based CMS because the protein level is much higher than
that of cane--approximately 18 percent for beet as opposed to
4 to 6 percent for cane (35).
V ILL-J2
-------
Animal feed supplements are widely employed in the raising
of cattle, sheep, poultry, and other domestic animals (40). These
supplements extensively utilize by-product materials such as
molasses, fat, and urea (40,100). CMS, derived from sugar cane,
sugar beet or citrus, may be substituted for part of the molasses
in some animal feed supplements (35T) . In this manner, not only
the CMS is utilized, but limited supplies of more expensive
molasses are extended (35).
CMS potentially could be employed in dry supplements and
liquid supplements. .There ar.e technical difficulties however,
in utilizing CMS in dry and block form supplements. Little CMS
(as a molasses substitute) can be used in feed blocks because-
technical problems related to the consistency of molasses have
severely limited its incorporation into solid blocks (40). There-
fore, it appears that utilization in liqu.id and dry feed supple-
ments offers the most favorable potential for marketing large
quantities of CMS.
Liquid supplements are available which contain a large per-
centage of molasses for high energy content. The molasses con-
tent of a finished liquid supplement may be as high as 70 percent
by weight (solids basis). The higher the molasses concentration,
the greater the stability of the liquid supplement (40). :.*
The trend in feed supplements has been away from dry sup-
plements to liquid forms because the latter are based on low cost
non-protein nitrogen sources such as urea instead of more costly
natural proteins such as soybean. Fluid materials also have cer-
tain advantages in blending, handling, and transport operations (100).
vrrr-i3
-------
Literature Review: Feed Acceptance and Nutritional Value
Feed Acceptance--
The results of some relatively recent animal feeding experi-
ments are summarized below.
A progressively severe condition of black and watery excre-
ment was noted in hens which consumed rations containing more
than 10 percent mostos. The appetite, weight, and appearance of
the birds remained healthy although, the watery excretions stained
eggs, hens, and surroundings (29).
CMS levels of 0, 1, 2.5, 5, 7.5, and 10 percent were incor-
porated into broiler starter diets (27). The eight-week body
^
weights of all birds receiving CMS were numerically superior to
those of the controls, with those of the 7.5 percent treatment
being th-e highest. Daily feed intake increased as the dietary
level of CMS was increased. Feed conversion efficiency was sig-
nificantly lower than the control for the 7.5 percent level of
supplementation (27).
Diets containing increasingly higher levels of CMS were also
fed to laying hens (27). Statistical evaluation revealed a sig-
nificant egg production improvement associated with the feeding.
of 2.5 percent or more of CMS. Egg weights and Haugh unit scores
(interior egg quality) were significantly reduced when 2.5 per-
cent or more of CMS-were fed. Specific gravity of eggs (shell
quality) and mortality were not significantly related to diet.
There was a trend toward increasing feed intake as CMS levels
were increased. The investigator concluded, on the basis of
VIII-14
-------
these data, that CMS appeared to be an acceptable and nutritious
product for use in diets of broilers and laying hens when proper
attention was given, during the diet formulation, to its overall
nutrient' content (27).
Under wire cage management practices, levels of cane molasses
as high as 20 percent can be incorporated in a practical diet for
laying hens (96). But the average poultryman in Puerto Rico has
his flocks under floor management, where molasses levels over 15
percent can cause a problem with, caked litter. Another factor
which would limit the use of locally-produced cane molasses in
commercial laying diets is its cost, when compared to that of
corn or other suitable more nutritive substitutes (96). Since
there is a similarity between cane molasses and CMS, this suggests
an upper lim-it of 15 percent CMS incorporation into poultry diets.
Dried mostos (10 percent moisture) was tested as a feed
ingredient (corn substitute) for laying hens (97). Above a level
of 20 percent, a serious detrimental effect on egg production
appeared. Average egg weight, hen weight, broken-open egg height,
and shell thickness were not affected by dry mostos levels as
high as 35 percent. This tended to disprove the theory that
high levels of potassium, magnesium, and sulfates in the mostos
r,
could have toxic effects such as the inhibition of calcium metabo^-
lism. Most of the-^aggs produced by hens consuming levels of dry
mostos above 15 percent were stained, and the animals themselves
were soiled. The overall data indicated that in practical diets
vrrr-i5
-------
for laying hens, the level of dry mostos should not exceed 15
percent (97).
The effects of feeding three percent dried molasses distil-
lers solubles to steers in all-grain rations was examined in
Australia (39).
Spray-dried mostos has been used in cattle nutrition as a
supplement in block or liquid form under grazing conditions.
Incorporation in the feed caused a decrease in animal live- and
carcas-weight gain and an inferior feed conversion ratio (39).
Forty Angus and Angus X Hereford yearling steers were
assigned to four treatment groups in a completely randomized
design for an 84-day feeding trial (2). Condensed molasses
solubles were substituted in a corn-soybean meal diet at 0, 5,
10, and 15 percent on a dry weight basi's. The CMS were obtained
from a cane-sugar molasses distillation operation. After 56
days, steers fed 15 percent CMS were changed to the control diet
due to reduced weight gain and feed efficiency. CMS at 5 and
10 percent also decreased average daily gain and feed conversion
efficiency, however, dry matter intake was not influenced by
treatment. At 84 days, there were no differences in average
daily gain or dry matter intake for steers fed 0, 5, and 10 per-
cent solubles, but feed conversion efficiency was reduced.
CMS had no effjact on carcass quality or yield. Chemical
analysis of CMS simples revealed high levels of potassium (11.6
percent, dry matter basis). CMS were judged to be le-ss valuable
than corn as an ingredient for finishing cattle. They were
vrrr-16
-------
judged as useful when maintenance of animal weight or less than
maximum gains were desired (2).
Nutritional Value--
The mean chemical composition of several cane CMS samples
is presented in Table VIII-1. The inorganic ash fraction comprises
approximately 30 percent of the dry mattar of the CMS. Protein
comprises approximately 9 percent. The potassium level is very
high. An amino acid analysis of cane CMS is also presented in
Table' VIII.-2 . This data is based on the mean of only two samples
but it compares, well with data presented elsewhere (68) for
"molasses ethyl alcohol fermentation solubles." Several of the
amino acids are present in relatively high concentrations.
The quality of molasses stillage as a feed supplement is
improved by desalting and replacing potassium with ammonium ion.
(94, 102). There are several variations in the desalting tech-
nique but generally sulfuric acid and gypsum are added to con-
centrated stillage at room temperature. The potassium content
is reduced by 70 percent and the-ash content in dry matter by
40 to 45 percent. The content of nitrogen-containing substances
is increased by 30 percent (65 , 172).
Potential for CMS Use in Puerto Rico and the U.S. Virgin Islands
The preceeding section has presented data indicating that "^
CMS can be successfully utilized when incorporated into animal
r*
feed rations at certain low percentages. This may be performed
by local feed manufacturers (on the islands) or, alternately,
CMS can be exported in bulk to one of several destinations -
VIII-17
-------
-TABLE VI.I'I-I. M£AN CHEMICAL COMPOSITION OF CANE CMS'
Component Concentration
Dry Matter (?) 52.56
Ash (%) 32.63
Crude Protein (*) 8.69
Ca (%) 2.06
P («) 0.16
K (%) 11 .59
Ma (%} 0.84
Mg (%) 1.07
Fe (ppm) 1865
Mn (ppm) 71 .7
Cu (ppm) 115.5
Zn (ppm 56.9
Dry weight basis (2) .
VI IT-18
-------
TABLE VILI.-2. AMINO ACID ANALYSIS OF CANE CMS*
Ami no Acid Concentration (%)
Aspartic Acid .791
Threonine .125
Serine .144
GTutamic Acid .403
Glycirve .118
Alanine .200
Valirie .159
Cystine (Cysteine) .026
Methionine .024
Isoleucine .101
Leucine .146
Tyrosine .072
Phenylalanine .086
Lysine .089
Histidine .046
Arginine . .069
Dry weight basi s (2 ) .
vrrr-i9
-------
United States, South America, etc. This discussion will only
consider the potential for local consumption.
Approach--
The potential for local CMS consumption is a function of
both the animal population and the amount of locally formulated
feed supplements. According to Soldevila the only likely vehi-
cle for incorporating CMS into the diets of local animals is in
feed concentrate prepared and sold by local commercial feed
mills (73). Therefore, estimates of the potential CMS market
t
must be dependent upon the quantities of feeds manufactured.
Since agriculture is virtually non-existent on any of the three
U.S. Virgin Islands, this analysis will be confined only to
Puerto Rico.
A degree of uncertainty exists relative to the maximum
amount of CMS that can be tolerated by cattle, swine,and poultry.
Many of the feeding studies cited in The Literature Review sec-
tion of this chapter have been conducted with, in some cases,
mostos (either dilute or spray dried) or with CMS. These tests
are not readily translatable into acceptable maximum or minimum
ration levels. For the purpose of.estimating CMS consumption,
therefore, several levels of incorporation were used - 5, 10, and
15 percent. Fifteen percent was chosen as the maximum allowabVe
level for feed incorporation as suggested by Dr. Soldevila's tes-
timony (73), in which he discussed homogeneity mixing problems
with liquids used as supplements to dry feed. While Or. Solde-
vila's remarks were directed toward molasses incorporation, it
vrrr-2o
-------
is assumed that CMS would offer similar mixing problems. Or.
Soldevila reports that the feed mills in Puerto Rico are only
incorporating five percent molasses into feed concentrate (73).
Therefore, to assume a percentage of more than five percent CMS
presupposes that the feed mills could change their mixing opera-
tions.
In 1975, a reported 423,057 T of animal feed were produced
loca-lly in Puerto Rico (73). Using this figure as a starting
point, Table VIII-3 presents the estimated maximum annual amount
of CMS which could be used at three differing levels of feed
incorporation. Estimates were based, in part, in the following
assumptions:
t All feed mills in Puerto R.ico would incorporate CMS
into their respective production
The. constituents contained in CMS are not harmful to
the consuming animals up to the 15 percent maximum
CMS is 60 percent solids, a percentage at which bacterial
activity is known to be negligible (32), and was con-
centrated from mostos at 8.35 percent total solids.
As can be seen from Table VIII-3, roughly 50. percent of the
total CMS produced could be consumed locally if utilized at
the 15 percent level. This is a hypothetical maximum. In %
reality, the total ^amount of CMS consumable through feed incor-
poration would probably be closer to the 21,000 T shown at the
5 percent level. This allows for both marketing and diet prob-
lems, surely to be encountered when initially utilizing a waste
by-product as a food supplement.
V111 - 21
-------
TABLE VI11-3. ESTIMATED MAXIMUM ANNUAL QUANTITIES OF
CMS WHICH COULD BE INCORPORATED INTO
PUERTO RICAN MIXED ANIMAL FEED*
Parameter
Ouantity'of feed mixed in
Puerto Ri'co
Quantity of CMS
supplement
used at 5%
Quantity of CMS used at 10%
supplement
Quantity of CMS used at 15%
supplement
Total annual quantity of CMS
(60% solids) produced by
Bacardi, Puerto Rico Dis-
tillers, and VIRIL
Incremental Subtotal
-Tons
Total
423,057
21 ,153
42,306
63,459
114,813
Based on 1975 feed figures (73)
-------
USE OF CMS AS AN INTERMEDIATE' PRODUCT
Introduction
The previous section discussed the use of CMS as an end
products in itself - i.e., as a potential animal feed supplement.
CMS, or a similar product with lower solids content, can also
be utilized as an intermediate product, from which certain by
products can be racoversd. Utilization of CMS as a intermediate
product in the following processes will be discussed.
Incineration - Control Parameters
Incineration - Equipment Options
Cost's of Evaporation/Incineration
'Potassium Recovery
t Organic Fertilizer Production
Control Parameters for Incineration
Incineration of mostos may be accomplished either directly
in the waste stream or after some degree of concentration. The
advantage of concentration is anticipated cost savings, on both
total equipment investment and operating costs. Engineers and
manufacturers of incinerating equipment have been quoted as say-
ing that it would be several times more expensive to incinerate
slops directly than to pre-concentrate and then incinerate (15).
Therefore, any cost-effective incineration plan is likely to "*
include an evaporation step.
«
Important control parameters which should be considered are
the incinerator feed solids concentration, the temperature in
vrrr-23
-------
the reactor, and the excess air supply. Calculations performed
on an analysis for a typical mostos stream indicate that auto-
genous combustion (combustion without the need for a continuous
supply of auxiliary fuel) can take place with a feed total solids
(TS) concentration as low as 35 percent (61). An analysis was
not performed to determine whether auxiliary fuel was needed to
initially bring the incinerator up to its normal operating tem-
perature, or whether the CMS could serve as start-up fuel.
A TS concentration significantly above 35 percent is desir-
able to increase heat recovery in a waste heat boiler (61) and
to decrease the size of the incinerator. A CMS concentration
of about 50 percent TS seems optimum since at higher concentra-
tions technical problems with evaporation conditions and
pumping characteristics of the CMS arise.(61). A further dis-
advantage of having a higher concentration is due to the fact
that the incinerator temperature must be maintained below 700 C
in order to prevent ash fusion. Pilot plant work (10) showed
that with a 75 percent TS feed, maintaining the incinerator
temperature below 700 C resulted in the need for higher excess
air conditions. In order to supply a large amount of excess
air, a larger sized reactor is required than would otherwise be
necessary.
The need for maintaining the incinerator temperature below'
,»
700 C is due to -the low melting point of the inorganic salts
contained in the waste. In general, the need to strictly con-
trol the temperature causes difficulty in attaining the correct
-------
balance of operating conditions for a suitable combustion tem-
perature with a complete combustion of volatiles (51). The dif-
ficulty of stabilizing the operating conditions is accentuated
by variations in the organic and inorganic composition of the
waste.
The variability in the waste composition is related to the
variations in the molasses raw material. Molasses is a by-product
of the sugar industry and is affected by variations in the ori-
ginal sugar cane source, in the sugar processing, and in the
sugar mill efficiency. Its characteristics, accordingly, vary
widely and cannot be completely predetermined or controlled by
the rum producers. Bacardi Corporation,, especially, because of
fts""Targe volume of.product, must rely on many molasses sources.
Several variations in types of incineration systems are
available, as each manufacturer tends to have their own innova-
tions. Few of these have been applied specifically to rum wastes.
However, wastes produced in the paper manufacturing industry
which are somewhat similar to rum slops, but containing sodium
rather than potassium salts, have been concentrated and fully
oxidized in a conventionally operated fluidized bed incinera- ..
tor (61). In 1915 rum slops were concentrated and incinerated
in the Porion Furnace and a system was later developed by Whita-,
>
ker (117) and U.S. .Industrial Chemicals, Inc. (4a , 43). During
<
the early 1940' s-'Reich (83) carried out experiments in low tem-
perature carbonization retorts at 343°C for the production of
carbon and potash. Further work- was carried out in the distillery
VI11-25
-------
of the Dyer Meaking Breweries Ltd., in Lucknow, India, in 1960
with a pilot-scale multiple hearth-type incinerator (1Q).
Kujala, et al. (61) have described an alternative to a
conventional fluidized bed incinerator, consisting of using the
fluidized bed reactor as a gasifier and performing the after
burning in an after combustion chamber which is linked to a
waste heat boiler. Apparently, the advantage of this type of
system over a conventional single reactor system is that it alle-
viates the difficulty of maintaining the proper balance of opera-
ting conditions for a suitable combustion temperature while
obtaining a complete combustion of volatiles. In the alterna-
tive fluidized system, the temperature in the gasifier is regu-
lated below the melting point of the ash, but is sufficient for
the volatilization of organic materials. Gases and volatilized
organics are separated from ash and are fully oxidized in the
after combustion chamber at a temperature which need not be
below the ash fusion point. Several incineration systems are
manufactured which utilize this same principle, having a primary
and secondary combustion chamber. Most of these however, are
not of the fluidized bed type, and therefore, do not require a
cyclone separator for ash separation. By providing near-complete
combustion of gases and organics in the secondary chamber, they'
have the ability to minimize air pollution. In fact, some of
them reduce polTutant emissions to the extent that the systems
can be installed without scrubbers. Combustion systems based
on this design are called controlled air incinerators.
VII1-26
-------
Costs of Evaporation and Incineration
The costs of evaporation alone have been discussed previously
under "Evaporation Technology." The same cost considerations
apply here. In addition, in a combined evaporation and inciner-
ation system there is an interrelationship between heat recovery
during incineration and fuel costs for evaporation. An attempt
to determine the costs of an evaporation and incineration system
depends to a. great extent on the accuracy of information with
regard to fuel consumption for both processes and on the poten-
tial for the recovery and use of heat energy in place of fuel.
Accurate information on the net energy requirements for evapora-
tion and incineration of rum slops is lacking.
George H. Oorion of Bacardi Corporation has evaluated pro-
ject designs for two types of incineration systems (34). The
first, the Copeland system, is a modified type of fluidized bed
system. It involves pre-concentration of mostos to 34 8rix.
During pilot tests with this method, auxiliary fuel was neces-
sary to sustain incineration. It was possible to incinerate
48 gal of slops/gal of fuel during the best test run. The.
investment cost for this system was estimated at $1,075,000 for
a unit processing 73,000 gal mostos/day. The operating cost/1,300
gal slops was estimated at $20.00.
A second system, the Prenco incineration system, was evalu
/*
ated and pilot tested. Being a liquid incineration system, it
required auxiliary fuel because of the low BTU/gal available
from mostos. The investment cost was estimated at $476,000- for
vr.rr-27
-------
a 33,600 gal mostos/day. The estimated operating cost/1,000
gal mostos was $24.40
Neither the Prenco nor the Copeland system designs were
based on maximum waste heat recovery, and neither included costs
for air pollution control or ash handling. Efficient waste heat
recovery would be expected to lower the net costs of treatment,
while air pollution control and ash handling requirements would
raise them.
Puerto Rico Distillers has estimated costs for evaporation
followed by a Brule Incinerator (15) which include costs for
ash handling and air pollution control, but again do not include
the savings possible with efficient, heat recovery. These costs
are much higher on a per gallons basis than Bacardi's estimates
for the Prenco and Copeland systems. This is suprising because
the former system involved production of CMS (50° Brix), while
the two latter systems did not. Puerto Rico Distillers has
stated that manufacturers who quoted them incineration equipment
costs agreed that it would be several times less expensive to
incinerate CMS than the dilute slops (15). Yet the estimated
capital expense of an evaporation and incineration system was
$4,655,000 for a 223,200 gal mostos/day, and the operating
cost was estimated at $47.00/1,000 gal slops. The corresponding
>
breakdown of costs ^.all ocated $400,760/yr for fuel oil for inci-
neration (of ato'tal direct cost for incineration of $579,440).
This fuel oil cost is based on 1.268 mil gal/yr at $0.32/gal.
This allocation is extremely high, amounting to 5,283 gal fuel
oil/day. In addition, $284,400 was allocated for fuel oil for
virr-28
-------
evaporation, or 3,750 gal/day. In total, 9,033 gal fuel/day
wars allocated for evaporation and incineration, equivalent to
25 gal slops treated/gal fuel. In contrast, the Copeland system
described earlier indicated that it was possible to treat a.s
much as 48 gal mostos/gal fuel. Puerto Rico Distiller's estimate
should be at least as low as 48 gal slops/gal fuel; and in fact
with the type of system proposed by Bacardi, fuel for incinera-
tion should be minimal except during start-up. While it IS not
possible to cal cul ate'the amount of fuel required, it is estima-
ted, that approximately 1,000 gal fuel would be a maximum require-
ment for incinerator start-up. Assuming that CMS is autogenous,
auxiliary fuel consumption to .sustain.combustion should be mini-
mal as compared to Puerto Rico Ois-tillers1 estimate, of 5,283
gal/day.
Potassium Recovery from Incinerator Ash
Evaporation and subsequent incineration is expensive in
terms of plant investment and operating costs. Therefore, it
is essential to obtain a return on the capital expenditure if
possible. The incinerator ash is itself a by-product which has
a potential value as a fertilizer component; however, by means
of further processing a more purified product may command a
higher market price. ::
*..
The incinerator ash is basically a crude potash containing
/*
potassium salts a'nd impurities such as calcium and magnesium
salts, silica, traces of iron and manganese, and unburned carbon-
aceous particles (61). Ash has been reported to contain 37 percent
vrrr-29
-------
potassium (expressed as K20) (59, 16). According to Kujala,
et al . (ST) potassium salts generally should comprise 65 to 70
percent of the ash. Expressed as a percentage of the ash, the
composition of salts is typically KpSCL, 56 percent; KC1 , 7.0
percent; and K-CO,, 5.0 percent. Some investigators (37) have
u 0
reported K2C03 contents of as high- as- 50 to 60 percent, with
only 15 percent K^SO,, and 10 percent KC1 .
The use of incinerator ash as fertilizer component, in con-
junction with nitrogen- and phosphorus-rich chemicals, is prac-
ticed at Izumi, Japan (24). A plant belonging to the Japanese
Ministry of International Trade and Industry producing non-
beverage alcohol incinerates some of its CMS. The ash is a fine
slate gray powder which is crushed in a mill and then sold to
a dealer. The dealer resells it as such or, more often, mixes
it with other materials to make a fertilizer. The plant repor-
tedly sells the ash for about $21/t and the dealer resells it
for about twice that price, or .still higher if it is converted
to a fertilizer.
Work on the recovery of potassium salts has taken place in
India (10, 61). A 90 gal/hr pilot plant consisting of evapora-
tion, incineration, and potassium salt recovery was constructed
at one of the Indian rum distilleries. Fresh slops were neu-
tralized with lime to pH 7.0, filtered in a filter press, and
fed to. an evapora-tor through a heat exchanger. CMS at 75 per-
cent dry solids was produced and fed to an incinerator where it
was ashed. The ash was removed and leached with water for the
vnr-3o
-------
dissolution of all the potassium salts. The solution was fil-
tered and neutralized with sulfuric acid to convert the carbo-
nates to sulfates. The neutralized liquor was concentrated in
an evaporator to 25 to 30 percent dry solids. Potassium salts
were crystallized out from the concentrate in a screw trough
crystal 1izer. A basket, centrifuge followed for separating sus-
pended crystals from magma. The crystals were then dried and
bagged, while the magma was reused in the process. The finished
product contained 83 percent potassium sulfate; 9 percent potas-
sium chloride with traces- of sodium salts; and 8 percent mois-
ture. A flow diagram of the entire potassium salt recovery pro-
cess is shown in Figure V'IIL-1 . .
Based on this research, it was estimated that a distillery
producing 72,000 gaT rum slops/day, could recovery 3.85 ton/day
of potassium salts (expressed as K20). At 1964 prices for
potassic fertilizers, the cost of the treatment plant could be
recovered in an estimated 5 to 6 yr from the sale of the by-
product (TO) .
Kujala, et al. (ST) state that the factory selling price
for a refined potassium salt by-product could be on the order
of $200/t within a few years. They report that there are indi-
cations that world resources of potassium salts, essential in ;*
N/P/K fertilizer formulations, are tending to reach a finite ~~
S"
limit in avai 1 abfl i ty, with anticipated future shortfalls in
supply. Kujala, et al. (61) state that potassium for fertili-
zer needs is normally sold as KC1 with a smaller market for
vrri-3i
-------
CRUDE POTASH
3-STAGE
LEACHING TANK
(COUNTERCURRENT
EXTRACTION)
DUPLEX
FILTERS
FILTERED
SOLUTION
NEUTRALIZING
TANK
NEUTRALIZED
FILTERED
SOLUTION
MAGMA
SHELL AND
TUSE HEAT
EXCHANGER
HOT
EVAPORATED
LIQUOR
SCREW
TROUGH
CRYSTALL1ZER
EVAPORATOR
MAGMA AND
SUSPENDED
CRYSTALS -
SOLIDS FOR
DISPOSAL
BASKET
CENTRIFUGE
SALT
CRYSTALS
DRUM
DRYER
CONDENSATE
STORAGE
AND
BAGGING
FINAL PRODUCT
.Figure VIII-1. Potassium. ..S,a\H Recovery From Crude Potash.
VIII-32
-------
K-SO, (the major potassium salt recoverable from rum slops).
With certain crops and in many tropical areas where soil sali-
nity levels are high, the chloride ion is undesirable and sul-
fate of potash is preferred. Quite often, these areas are to
be found where sugar cane is grown.
The market for potassium salts available to the Puerto
Rican and Virgin Islands rum producers is uncertain. Table VIII-4
presents estimated potassium production figures based on the
following assumptions:
Incinerator ash is produced at the rate of 0.2 Ib ash/
. gal of 8.35° Brix mostos
Ash is 55 percent total potassium salts (37 percent
expressed as. K-Q) .
Qf the total estimated .annual production of 15,000 T, 5,550 T
(expressed as 1^0) would be available for potential use as a
fertilizer source. To place this figure in proper prospective,
the potential local market for potassium needs must be examined.
Table VIII-5 presents the annual average consumption of
potassium chloride and other potassium salts by direct applica-
tion (not mixed with other nutrients) in Puerto Rico over the
period 1965 to 1975 (109) .
vrrr-33
-------
TABLE VIII-4. ESTIMATED ANNUAL POTASSIUM PRODUCTION
AS AN INCINERATION BY-PRODUCT
i
Dally Generation of
Mostos (gals/ day)
Operating days/yr
Ash Production (T/yr)
Potassium Production (T/yr)
(Expressed as K^O)
Remaining Residue to
be Disposed (T/yr)
Bacardi
300,000
300
9,000
3,330
5,670
Puerto Rico
Distill ers
200,000
200
4,000
1 ,480 .
2,520
VIRIL
100,000
200
2,000
740
1 ,260
Totals
600,000
--
15,000
5,550
9,450
-------
TABLE VIII-5. PUERTO RICAiN ANNUAL COMMERCIAL
CONSUMPTION OF POTASH AND
POTASSIUM COMPOUNDS IN DIRECT
APPLICATION MATERIALS*
Potassium Product Annual Average Consumption (Tons)'
Potassium Chloride, 50-62%
Grade (expressed as KC-1 ) 1,110
Other Forms 586
Total 1,696
*' (109)
1 Based on 1965 to 1975 consumption figures.
Assuming that the bulk of KC1 is in the normal 60 percent
grade, the 1,110 T would be equivalent to 802 T (expressed as
K^O). Similarly, if the "Other Forms" are predominately K^SO^,
the 586 T could be translatable into 318 T expressed as K-0 .
The estimated <20 total of 1,120 T (802 +313) can then be com-
pared to the hypothetical production total of 5,550 T (Table
VIII-1). As ,can be seen, production would exceed current use
by roughly five times. Assuming that local consumption would not
increase, the difference would have to be exported at a presumed
price disadvantage. -^
Organic Ferti 1 i zer "Production from.CMS
Molasses alcohol stillage contains considerable amounts of
organic materials including amino acids, proteins, saccharides,
gum materials, etc. Kyowa Hakko Kogya Co., Ltd., of Tokyo and
vrri-35
-------
Hofu, Japan, has developed a process for creating an organic
complex N-P-K fertilizer from stilla.ge by converting the organic
components to humic acid (24). The humic acid forms an organic
complexing agent to which inorganic nutrients are coupled. This
coupling controls the release of the nutrients for plant growth.
Consequently, the damages caused by high concentrations of nutri-
ents often seen in inorganic chemical fertilizers are minimized.
Kyowa Hakko Kogyo Co. obtains sugarcane molas.ses in Japan
and from Indochina, the Philippines, and Taiwan. The cane
molasses is converted into alcohol and is used in the production
of Pharmaceuticals and other products at Hofu, Japan (24). The
company has performed experiments (44) which revealed signifi-
cant damage to germinating plants when untreated cane molasses
stillage was applied to the soil. However, if the stillage was
treated with mineral acids, the damage was essentially elimina-
ted. The damage was apparently caused by organic acids contained
in the untreated stillage or by the acids formed in the soil
from the sugar component of the stillage. By treating with min-
eral acids, the low molecular weight organic compounds were
dehydrated, condensated, and converted to humic acid.
A procedure for producing fertilizer from stillage involv-
ing evaporation to CMS prior to subsequent treatment steps was >
developed. A pilot program using a plate-type triple effect
evaporator with
-------
With five of these evaporators, the Hofu plant has produced
approximately 140,000 t/yr fertilizer (24). The evaporators
produce CMS with approximately 50 percent total solids. In the
fertilizer plant, sulfuric acid is added to adjust pH. The mix-
ture Ms then heated to about 110°C for 8 to 10 hr as it is
passed through a series of six - 300 gal reactors. Ammonia is
added to achieve a pH of 4 to 5. Finally, phosphoric acid and '
potassium sulfate- are added to adjust the composition of the
final fertilizer. The resulting mixture has a soupy consistency.
An agglomeration process then follows, resulting in a granulated
but moist product. The granules are dried in a direct-fired
(with heavy oil) rotary kiln which is about 6. ft in diameter
and 70 ft long. Treatment of the exhaust gasses from the rotary
kiln is required. The dried granules are cooled and put through
a vibrating sifter to obtain the final product. A flow diagram
for this process, is shown in Figure VIII-2.
The final product' has the desirable characteristics of
being homogeneous, granulated., and available in several grades.
It has been commercially sold for $212 to $220/t in Japan. A
lUS. selling price of $160/t was proposed (24). Kyowa Hakko .
Kogyo Co. advised that this type of fertilizer production was
not a break even proposition unless annual quantities were in v_
excess of 10,000 t/yr (24). They estimated that if VIRIL were to
employ this process, their annual production of fertilizer would
be approximately 7,000 t. Assuming that this estimate was based
on a yearly production figure of approximately 3.4 million proof
Viri-37
-------
I
' CJ
CD
FERMENTATION
WASTEWATER
[CA- SUPER PHOSPHATE j [H3PQ4 | |KgSQ4j j NH3;
*REF. 58
A
B
C
D
E
EVAPORATOR F
GLASS LINED REACTOR G
NEUTRALIZER H
KNEADER I
MIXER J
GRANULATOR K
ROTARY DRYER L
FIRST ROTARY COOLER M
SIFTER N
CRUSHER O
PRODUCT
SECOND ROTARY COOLER
HOPPER
WASHING SPRAY TOWER
WET-TYPE COTTRELL DUST COLLECTOR
AFTER BURNER (INCINERATION)
.figure VIII-2. Process Flow Diagram for the Production
of An Organic Compound Fertilizer*
-------
gallons, organic fertilizer estimates for Bacardi (.17.4 million
proof gallons) and Puerto Rico Distillers (5.2 million proof
gallons( would be 35,800 and 10,70.0 t, respectively.
VIII-39
-------
IX. POLLUTANT REDUCTION THROUGH
PROCESS MODIFICATIONS
There are several process modification options available to
the rum distillers which, if implemented, would effect partial
reductions of certain pollutants. Those of significant
importance are:
t Pretreatment of molasses prior to fermentation
» Removal and recovery of yeast from fermenter beer
Removal of fermenter bottoms
Recovery of fusel oils and "heads" as a fuel
supplement.
These are all options which the permittees have indicated that
they could implement quickly, i.e., within 6 mo to 1 yr (80).
Unfortunately, together, these methods are expected to give a
maximum BOD5 reduction of only 13 percent and TSS removals of
slightly more than 65 percent (80). The SOOg reduction is so
low because only the insoluble solids are affected by these
methods, and approximately 80 percent of the waste 800^ is in
soluble form (93). Nevertheless,, these options may be considered
as feasible because of the relative ease of implementation, the
possibilities for recovery of valuable by-products, and the posi-
tive effects on subsequent mostos treatment options.
'£*
PRETREATMENT OF MOLASSES
Molasses pretreatment clarifies and pasteurizes the raw
molasses, removing gums, sludges, salts, and unwanted micro-
organisms. These components tend to inhibit the yeast activity
-------
and, if not removed, will become unfermentable solids in the
bottoms of the fermentation tanks (61).
Molasses pretreatment may be achieved by diluting molasses
to approximately 40 Brix and raising the temperature close
to the boiling point, when, due to the inverted solubility of
the calcium salts, mainly calcium sulfate, maximum sedimentation
occurs. Under these heated conditions, separation can be
carried out in large settling tanks or by use of an ejecting-
type molasses clarifier. After clarification, heat recovery
is possible by exchange with the incoming, freshly-diluted 40
Brix molasses (61 ).
Molasses pretreatment has important side benefits. Clarifi-
cation and pasteurization of the molasses substrate yields a
cleaner alcohol; consequently, the purifying column in the
distillation unit requires fewer stoppages for descaling, and
reduction in the build-up of sludges in the boiling column per-_
mits 'longer operating runs. Most importantly, from the stand-
point of effluent treatment, the slops coming from the stills
are cleaner and would be expected to cause less scaling of
mechanical treatment equipment,, e.g., evaporation units (61).
However, there is a large energy committment and capital
expense involved in molasses pretreatment (80 , 67 ). In addition-',
there is reportedly, some 1 oss of fermentable sugars ("80). This
is more of a problem as molasses becomes more expensive, there
being a trade-off between benefits gained from pretreatment
versus sugar losses (decreased rum yields).
IX-2
-------
A partial .pretreatment of molasses can be effected simply
by dilution and then centrifugation. Some distillers prefer
this method, which does not involve heating, or forced
sedimentation (67, 80). Without heating, however, neither
pasteurization of the molasses nor precipitation of calcium
salts is achieved. Figure I'X-1 presents a typical molasses
centrifugation operation.
YEAST RECOVERY
Yeast may be eliminated from the fermenter beer by
centrifugation (Figure I'X-2). The centrifuged yeast may then
be recycled for further use in the fermenters, sold for an
animal feedstuff (with or without drying), or land disposed.
The Bols Distillery in Scheidam, The Netherlands, practices
yeast centrifugation and recycling "(24). Yeast used in the
fermentation is removed from the. beer by a Westphalia centrifuge
and is recycled. The separated yeast is acidulated with
hydrochloric acid, to a pH of 1.8 to 2.0 to prevent spoilage
during storage. When needed again for fermentation, sodium
hydroxide is added to produce a yeast cream of pH "5, which
reportedly has a total solids content of 10 percent and contains
45 percent protein and 3 percent ^2^5* Urea and phosphorus are
added as nutrients; fermentation is accomplished in about 12 hr/
'
The yeast is continuously recycled for 3 to 4 mo, at which time
it is discarded to the local sewer system (24).
IX-3
-------
CONTINUOUS IN-LINE
MIXING
MOLASSES
WATER
DILUTED MOLASSES
STORAGE TANK
CENTRIFUGAL
SEPARATOR
L
LAND DISPOSAL
SLURRY PUMP
SOLIDS RECOVERY
TANK
CENTRIFUGED
MOLASSES STORAGE
TANK
PUMP
TO
FERMENTATION TANKS
Figure IX-1 , Molasses centrifugatlon (80)
-------
WATER
x
en
PUMP
7
YEAST CREAM '-YEAST CREAM
RECEIVING TANKS WASHING TANKS
YEAST
CENTRIFUGAL
SEPARATOR
PUMP
- YEAST
CENTRIFUGAL
SEPARATOR
FERMENTATION
TANKS
.HYDROCHLORIC
ACID
YEAST
CREAM
* HOLD.ING
' TANK
PUMP
DISPOSAL
~OF DILUTE
SIDESTREAM
YEAST CREAM
TO PROCESS
BEER TO DISTILLERY
CLARIFIED BEER
STORAGE TANK
F1 gure/. I-X-2. Yeast recycling by centri fugation of fermenter beer (80).
-------
The advantages of yeast recycling are reported to
include:
Shorter fermentation times (61)
Increased alcohol yield (61)
Greatly diminished demand for new bakers' yeast
Short technology implementation time (80)
Less clogging of evaporation equipment used for
treatment of mostos (24)
Reduction of the amount of yeast requiring disposal
to a small amount that is wasted (80)
Suitability of the wasted yeast for land application
or use as an animal feedstuff
Availability of an alternative option for by-
product recovery: sale of yeast as a high-
protein animal feed supplement.
Possible disadvantages of yeast recycling include:
Requirement for molasses pretreatment (6T , 67)
Unknown and perhaps undesirable effects of molasses
. pretreatment on alcohol yield and/or quality
(67, 80)
Production of higher temperatures in the fermenter,
possibly necessitating installation of a fermenter
chiller (80.)
Requirement of land disposal of molasses pretreat-
ment residue and wasted yeast.
A particular version of yeast recycling is known as the
Melle-Boinet technique. With this technique, fermentation
times can be shortened to 14 hr, with up to a 5 percent increase
in alcohol yield, according to Kujala, e_t_ ajk (61); without
yeast recycling, fermentation times can take 24 to 72 hr (36).
IX-6
-------
With this technique, a high yeast concentration is maintained
in the fermenter and cell reproduction is almost eliminated,
transforming sugars previously used for yea'st growth into
alcohol. Providing the yeast is handled properly and biological
infections in the fermentation are controlled, it can be
recycled up to 30 times before an appreciable loss in activity
occurs ( 61).
Use of the Melle-Boinet technique in industrial spirits
manufacture can only be made where molasses has first been
pasteurized and pretrsated to remove components which inhibit
the yeast (61). Therefore, there is the risk that sugar losses
will offset any gain in alcohol production from increased
efficiency of sugar utilization.
REMOVAL OF FERMENTER BOTTOMS
Distilleries have the option of eliminating suspended
inorganic and organic materials (other than the live yeast) which
could become part of the rum slops in either of two ways:
by molasses pretreatment, which has already been discussed, or
by the removal of fermenter bottoms.
Fermenter bottoms are the unfermentable solids and dead
yeast cells which settle to the bottom of a fermentation tank.
The sludge is both inorganic and organic, being derived mainly
from the impurities" of the original molasses. Fermenter bottom
?
sludge is minimized if molasses pretreatment is practiced;
otherwise, several hundred gallons per fermentation tank of a
high BOO sludge is formed (61). When fermentation ceases, the
IX-7
-------
liquid containing the live yeast cells is drained off for
yeast recovery and/or distillation and the bottom sludge must
be washed out of the tanks before they are reused for fermenta-
tion. A schematic flow diagram which shows how fermenter
bottoms are removed is included as Figure IX-3.
It has been the practice at many distilleries to either
run this material down a sewer or to dispose of it with the
slops. An alternative is removal via a sludge pump to a holding
tank and then land disposal (80). It is generally accepted that
this material cannot be treated with the rum slops due to its
effect on mechanical equipment, especially evaporation
equipment (15, 24).
The advantages of removal of fermenter bottoms over
molasses pretreatment are reported to be lower capital and
operating (energy) costs and higher alcohol yields (80, 67).
If yeast recycling is practiced, however, alcohol yields may,
in fact, be higher wi'th molasses pretreatment (61).
RECOVERY OF FUSEL OILS AND "HEADS"
The "heads" are the aldehydes and esters which are drawn
off the top of the second distilling column, the aldehyde column.
The fusel oils and amyl oils are separated in a third column,
the rectifying column. To recover these streams as fuel, they
are pumped from the"distil1ing columns to a storage tank and then
r
to a second burner installed in the plants' boiler.
IX-8
-------
FERMENTERS
X
I
10
L
WASHOUT
PUMP
BEER
DUMP
CENTRIFUGE
f
SLUDGE
PUMP
SOLIDS
TANK
YEAST
TANK
SURGE
TANK
SOLIDS TO
TRUCK FOR
LAND DISPOSAL
YEAST TO
TRUCK FOR
SALE OR
LAND DISPOSAL
r8EER
PUMP
TO
BEER
STILLS
JX-3. Removal of fermenter bottoms and yeast (80).
-------
Both VIRIL and Puerto Rico Distillers already recover
fusel oils and heads and burn them as fuel on a continuous
basis. The heads and fusel oils would otherwise become part
of the waste load requiring treatment.
COST
Cost estimates from Puerto Rico Distillers and VIRIL (80)
for some of the options which have been discussed are presented
in Table IX-1. Estimates were available for the removal of
fermenter bottoms, centrifugation of molasses, yeast recycling
(without molasses pretreatment}, and burning of fusel oils and
heads. Cost breakdowns for each are given in Appendix IX-11
(80). Except for the estimate for oil burning, none of these
estimates incorporate any cost savings due to by-product
recovery, or to positive impacts on rum production or waste
treatment. Such savings could be considerable.
ix-io
-------
TABLE IX-1. COST ESTIMATES FOR POLLUTANT REDUCTION THROUGH PROCESS MODIFICATIONS,
...FROM PUERTO RICO DISTILLERS AND VIRFC*' (80)
Puerto Rico Distillers VIRIL
Process CapitalAnnualCapital Annual
Removal of Fermenter $50,000 $38,260 $58,000 $26,840
Bottoms
Yeast Recycling by $200,000 $77,150 $232,000 $87,660
Centrlfugation
Molasses Centrifugation $110,000 $59,585 N/At N/A.
Burning of Fusel Oils $15,000 $0# $11,500 $0
and Heads
^Submitted January 30, 1978-
tN/A = Not available.
^Assumed to be offset by fuel savings.
IX-11
-------
TABLE IX-2. BREAKDOWN OF COST ESTIMATES FOR POLLUTANT
REDUCTION THROUGH PROCESS MODIFICATIONS, *
. FROM PUERTO RICO DISTILLERS AND VIRIL. (80)
REMOVAL OF FERMENTER BOTTOMS (VIRIL)
Capital Costs
Solids tank $15,000
Pumps, piping and electrical $15,000
Structural $8,000
Dump truck $20,000
Total $58,000
Annual Operating Costs
Depreciation $5,800
Maintenance $1,740
Labor $12,000
Electricity $1,500
Financing $5,800
Total $26,840
REMOVAL OF FERMENTER BOTTOMS (PUERTO RICO DISTILLERS)
Capital Costs
Pit tank $2,000
Slurry pump $3,000
Thickeners $10,000
Water storage tank $5,000
Water pump $1,500
Piping, rearrangment of fermenters, etc., $3,500
Dump truck $20,000
Electrical installation, etc. $2,000
Contingencies $3,000
Total $50,000
Annual Operating Costs .(Direct)
Transportation and dumping of decanted
slurry $14.400
Subtotal $14,400
*Submitted January 30, 1978.
IX-12
-------
Table IX-2 (.continued)
Annual Operating Costs (Indirect)
Repairs and maintenance $2,500
Property tax $1,200
Truck license $400
Depreciation
Truck $4,000
Other equipment $3,000
Insurance $500
Payroll benefits (402 in Statutory
and corporate benefits) $5,760
Financing $4,500
Contingencies $2,000
Subtotal $23,860
Total $38,260
CENTRIFUGATION OF FERMENTER BEER (PUERTO RICO DISTILLERS)
Capital Costs
Centrifuges (2) $120,000
Centrifuged mash receiving tank $18,000
Centrifuge feed pumps (2) $6,000
Yeast cream receiving tanks (2) $5,000
Yeast cream pump $3,000
Yeast washing tank $3,000
pH controller $4,000
Piping, valves, electrical, installa-
tion, etc. $10,000
Engineering consulting fees $25,000*
Contingencies $6,000
Total $200,000
Annual Operating Costs (Direct)
Labor (1 laborer) $8,280
Chemicals $4,000
Electric power $S,QQQ
Subtotal $17,280
Annual Operating Costs (Indirect)
Repairs and Maintenance" $5,000
Property Tax '*" $7,860
Depreciation $20,000
Insurance $700
Payroll Benefits (40% in statutory
and corporate benefits) $3,310
IX-13
-------
Table IX-2 (continued)
Financing $18,000
Contingencies $5,000
Subtotal 559,870
Total $77,150
*For study of yeast recycling following centrifugation.
CENTRIFUGATION OF FERMENTER BEER (VIRIL)
Capital Costs
Centrifuges (2) (Sharpies Model P-3400) $160,000
Electric generators $50,000
Pumps, piping and electrical $TO,000
Structural $12,000
Total $232,000
Annual Operating Costs
Depreciation $23,200
Maintenance $4,860
Labor $6,000
Electricity $30,400
Financing $23,200
Total $87,660
CENTRIFUGATION OF MOLASSES (PUERTO RICO DISTILLERS)
Capital Costs
Molasses diluting $5,000
Mixer, in line $5,000
Centrifuge $60,000
Solids recovery tank ' $2,000
Centrifuged molasses tank $5,000
Pump for 60° Brix molasses $3,000
Piping, valves, eletrical installation,
etc. $4,000
Dump truck $20,000
Slurry pump $3,000
Contingencies $3,000
Total $110,000
Annual Operating Costs (Direct)
Transportation and dumping slurry $14,400
Electric power . $3,000
Subtotal $17,400
IX-14
-------
Table IX-2 (continued)
Annual Operating Costs (Indirect)
Repairs and Maintenance $3,000
Property tax $4,325
Depreciation
Truck $4,000
Other equipment $9,000
Insurance $800
Truck license $400
Payroll Benefits (40% in statutory
and corporate benefits) $5,760
Financing $9,900
Contingencies S5,000
Subtotal $42,185
Total $59,585
BURNING OF FUSEL OILS AND HEADS (PUERTO RICO DISTILLERS)
Capital Costs
Storage tank $5,000
Pump $1,500
Burner $3,500
Piping and installation $3,000
Other $2,000
Total $15,000
BURNING OF FUSEL OILS AND HEADS (VIRIL)
Capital Costs
Storage tank
Burner
Pump and piping
Total
$5,000
$2,500
$4?000
$11,500
IX-15
-------
X. AQUACULTURE
INTRODUCTION
The fermentation and distillation of rum from cane molasses
produces relatively large quantities of hot acidic wastes, high
in BOO, COO, solids, and Kjeldahl nitrogen (80).. Presently, these
wastes are disposed of untreated to inshore waters off Puerto
Rico and St. Crorx, Virgin Islands. In an effort to develop a
satisfactory solutf'on to the problem of rum slops disposal,
the rum distillers have given more attention to evaluating aqua-
culture as a method for utilizing these .wastes. Aquaculture has
the advantages of requiring little or no pretreatment of the
effluent, considerable potential for recycling nutrients from
the waste into a harvestable crop, and providing acceptable levels
of predischarge treatment.
In concept, the advantages of aquaculture are numerous;
however, several obstacles stand between the concept and its
practical application. Currently, there is limited knowledge .
of the manner in which rum slops can best be used in an aquacul-
i
ture project, and what species or chain of species leading to a £.
harvestable crop ca-n utilize the slops efficiently. Questions
as to the physical facility requirements, whether these facilities
can be located in reasonable proximity to the rum distilleries,
and what the projected capital and operating costs would be are
still unanswered.
X-V
-------
PREVIOUS RESEARCH IN MOSTOS UTILIZATION B,Y AQUATIC ORGANISMS
The industry has solicited proposals for research into mostos
utilization in aquaculture in an effort to answer these questions.
A preliminary proposal for a feasibility study was received from
Or. Colin E. Nash, Oceanic Institute, Makapuu Point, Waimanalo,
Hawaii (80). A draft proposal was also solicited and received
from Or. Oswald A. Roels, University of Texas Marine Science
Institute, Port Aransas Marine Laboratory (80). Dr. Roels has
previous experience with an aquaculture project on St. Croix,
although this earlier study did not involve mostos. Dr. Roels
has proposed to analyze the nutritive value of slops which will,
in turn, aid in the selection of an organism or organisms suitable
for treating and recycling each component of the waste. Selec-
tion of suitable organisms will be based on the following cri-
teria:
Potential for removal of dissolved inorganic nitrogen
and/or phosphate
Removal of particulate protein and carbohydrate
Removal of soluble protein and carbohydrate through cul-
turing of unicellular heterotrophs. These, in turn,
may be harvested and used directly as a protein source
or fed to a higher trophic level aquatic organism.
Determination of tolerance levels to various concentrations
/*
of mostos and the^ability of test organisms to grow and provide
satisfactory waste treatment will also be investigated. Finally,
Or. Roels''s study wou'ld attempt to provide preliminary data on the
input requirements and potential yields of aquaculture systems.
X-2
-------
The University of Puerto Rico is actively involved in aqua-
culture research. The principal objective of their studies has
been to discover additional sources of protein-rich foods for -
human consumption in Puerto Rico. Most of the current research
concerns the growth of the fresh-water fish, Tilapia aurea. Only
one study utilizing mostos wastes has been conducted. This study
investigated the affects of mostos on Tllapia aurea under culture
conditions and on the water quality of the culture system (80).
A method for efficiently breeding and culturing fish, shell-
fish, and zooplankton with waste cane molasses has been developed
by Kawano,. et al (53). Applicability of this research to potential
mostos aquaculture experiments, however, is questionable due to
an inadequate description of the waste molasses used in Kawano,
et al,. studies. Characteristics of the waste molasses and its
similarities- to mostos are not provided in the patent information
filed by the authors (53).
Tosteson, et al (107) have conducted numerous studies with
mostos to determine its effect on marine organisms. Their work
examined four components of mostos:
Filtered crude mostos suspension (Fraction FT)
Crude mostos dialyzed against distilled water (Fraction '
F2)
. ;5t
Precipitate formed by the addition of three volumes of
i*
cold ethyl alcohol to crude mostos fraction (Fraction F3)
Fraction of crude mostos soluble in cold ethyl alcohol
(Fraction F4).
X-3
-------
In their experiments, they used cultures of the marine algae
Chlorella vulgar is, and isolated the marine bacterium Photobac-
terium fi scher from sea water. Their results indicated the fol-
lowing:
Fraction Fl appeared to promote the growth of marine
bacteria while inhibiting that of micro-algal cells. The
crude mostos fraction may have inhibited the algal growth
by acting directly on the cells or by simply promoting
the growth of bacteria in the algal suspensions.
Nondialyzable high molecular weight components of Fraction
2 promoted the growth of algae and inhibited bacterial
growth. This suggested that the alcohol-treated material
(Fractions 3 and 4) promoted both the growth of the algae
and the bacteria.
r The inhibitory effect of Fraction Fl on algal growth.
appeared to be due to the presence of low molecular weight
components which were removed by dialysis. These dialy-
zable materials at high concentrations appeared to stimu-
late bacterial growth while obscuring the inhibitory
effects of high molecular weight material.
The crude mostos fraction (Fl) contained materials of.
high molecular weight that were stimulators to micro-
algal growth; however, the presence of dialyzable low
<
molecular^weight substances in this same fraction promoted
greater bacterial growth which, in turn, restricted or
inhibited algal growth.
X-4
-------
POTENTIAL FOR UTILIZATION OF NUTRIENTS
SCS Engineers contacted two noted authorities in aquacul-
ture. research for information on the potential for utilization of
mostos in crustacean and higher organism aquaculture systems; Or.
Douglas Conklin, Associate Director of the Aquaculture Program,
Bodega Marine Laboratry of the University of California, and
Dr. Michael Hartman, Division of Natural Sciences, Savanah State
College, Georgia. Copies of their correspondence are included
in Appendix A. Both agreed that the mostos material was not
adaptable as a feed ingredient in aquaculture systems. Their
evaluation was based on several factors including the following:
The 800 and COO are very high indicating that the mostos
is very high in organic carbon compounds.
High carbon content is further substantiated by high car-
bohydrate content.
Presence of significant amounts of carbon compounds will
enhance growth of heterotrophic bacteria. These bacteria
can create situations of lowered dissolved oxygen, as
well as destroy the bacterial flora of a biological fil-
ter system,should a system of this type be in use as
part of the treatment process.
Levels of copper in mostos (average 32.8 ppm) are toxic \
to most sheJlfish and acutely toxic to the larval stages.
i»
Potential^ for bioaccumulation of copper and other metals
by aquatic organisms may be considerable considering con-
centrations available.
X-5
-------
Although additional research is needed in this area, the
information currently available in the literature and from indi
viduals active in this research field suggests that mostos is
unsuitable for aquaculture systems.
X-6
-------
XI. BIOLOGICAL TREATMENT SYSTEMS .
INTRODUCTION
Biological treatment of rum slops has long been a proposed
solution to the rum waste disposal problem (11). However, the
consistently high 800^ has-'not favored aerobic treatment, and the
effluent from anaerobic treatment has been demonstrated to be high-
in B005 (2,000 to 10,000 mg/i). Consequently, research has been
directed at the use of combinations of biological systems (mainly
anaerobic followed by aerobic processes) for the treatment of rum
slops-.
AEROBIC TREATMENT OF MOSTOS
A review of the literature, indicates that the application
of conventional aerobic processes (activated sludge or trickling
filters) is beset by problems. In general, the literature reports
that to be amenable to aerobic treatment the raw mostos would
first have to be cooled to a temperature range conducive to meso-
phiMic organisms (25 to 35°C), neutralized to pH 7.0-8.0, and
have supplemental nutrients added (phosphorus and/or nitrogen).
Sen, et al . (92), reported that trickling filter treatment
of undiluted molasses distillery waste was found to remove 75
/*
percent of the BOO",-. A large filter area was needed due to
the high concentration of 8005 in undiluted slops (18,000 to
50,000 mg/2.) and the low 800g loading rate that can be applied
XI-1
-------
to the filter. The process was dismissed as inapplicable because
of the space required by the filters.
Burnett (19) reported that the activated sludge process was
unsati sfactory -when used with a 10 percent solution of rum slops
in domestic sewage. The maximum BODr removal observed was 28 per-
cent. Also, difficulties were caused by extensive foaming. How-
ever, the investigator found that the activated sludge process
was highly efficient for the treatment of one percent slops in
domestic sewage. Over 90 percent efficiency in BQOj. and suspended
solids removal was observed.
Kujala, et al . (61), estimated that in 1972 the cost for oper-
ating an activated sludge treatment plant for an incoming slops
stream of 35,000 mg/4 8005 at a flow of 1,000 m3/day (0.26 MGD)
would exceed $2.5 million per annum. Space requirements for the
installation and the electrical loads for the necessary aeration
equipment were estimated to be large. Consequently, the authors
rejected activated sludge treatment of the raw slops on economic
grounds.
ANAEROBIC TREATMENT OF MOSTOS
Anaerobic digestion has been widely investigated as a method
for the treatment of molasses distilling mostos, and has been
found to be a partially effective treatment alternative. Anaerob'ic
v
digestion has the ability to handle the raw undiluted slops without
the problems associated with aerobic biological treatment, e.g.,
sludge bulking. Various designed and operational methods for
anaerobic digestion of mostos hava been researched. The findings
will be presented on the following pages.
. XI-2
-------
Sen, et al ., conducted experiments on laboratory scale
anaerobic digestion of molasses distillery mostos 'in India (92).
They found that it was possible to carry out anaerobic digestion
of the waste at 37 C using a cow manure seed at a BODe loading
of 3.0 kg/m /day with a retention period of 10 days. The diges-
tion gave an off-gas to waste volume ratio of 25:1, with an
overall efficiency of 0.86 m of gas per kg of S00g digested.
The gas was 60 percent methane. SOOj removals of more than
90 percent were obtained with a BOD5 reduction from about
30,000 mg/2, to less, than 1,000 mg/i. Digester failure was exper-
ienced at a loading of 3.8 kg/m /day with a. retention time of
eight days. Nutrient addition was found to be unnecessary.
Recirculation of CO^ after separation from other constituents of
the digester gas was more effective than mechanical stirring or
recircul atio-n of the entire digester gas in improving BOOg and
volatile solids reduction and the percentage of methane in the gas.
Shea, et al., in a study for the EPA investigated an anaerobic
digestion pilot plant for the determination of design criteria for
the full-scale application of anaerobic treatment and to estimate
the full-scale costs for such a process (93). The pilot system '!
gave a COD removal efficiency of 57 to 70 percent at a solids
retention time greater than 40 days. The influent mostos stream;
contained 70,000 to 100,000 mq/i. The range of total annual
i*
costs (including amortization and operating and maintenance costs)
was estimated to vary from $3.74/m treated at a design flow of
190 m3/day to $2.13/m3 treated at a design flow of 1,140 m3/day
(January 1974 cost levels). The recovery of methane gas as an
XI-3 :
-------
energy by-product of the anaerobic treatment was estimated to
reduce the above-mentioned unit treatment costs from one-third
at the 190-m3/day flow rate to two-thirds at the 1 ,140-m3/day.
flow. It was also concluded that additional treatment would be
required to meet the proposed effluent quality levels as defined
in the distillery's respective NPOES permits.
Basu and Leclerc compared the treatment of beet molasses
distillery mostos by thermophilic and mesophilic anaerobic
digestion (9). This research determined that gas production
was less consistent in conventional digestion when compared to
units with the thermophilic or mesophilic digestion. This was
attributed to the loss of gas during settling and entrained in
effluent solutions. Digester loading had no direct bearing on
gas production, and instead was related to retention time. The
efficiency of rate of methane production during high rate
digestion was related to C02 recirculation, as shown by Sen and
Bhaskaran (92), or to efficient mechanical stirring.
All the experiments were done with 10 days retention time
at varying BOD loadings of 2.0, 2.5, 3.2, and 3.5 kg/m /day.
The thermophilic digestion BOD removal for the first three loadings
were 97.2, 95.7, and 87.5 percent respectively, as against
corresponding removals of 96.0, 96.4, and 95.9 percent BOO >
removal for mesophilic digestion. Both thermophilic and-meso-
s-
philic digestion exhibited digester upset at the 3.5 kg 80D/m /day
loading. Basu and Leclerc concluded that, taking all aspects
into consideration, thermophilic digestion performs as well as
XI-4 ,
-------
mesophilic digestion, but the process yields only minor benefits
at the expense of higher costs needed to maintain the digester
at 55°C rather than 35°C (9) .
Hiatt, et al. , found that the. start-up and initial acclimati
zation period was very significant in the anaerobic treatment
of mostos (4>8)-r They determined that nearly three months
were required for the biological system to acclimatize itself
to a new substrate. After stabi1ization, a change in the loading
rate required, 10 to 15 days, for a return to normal operating
conditions. Dilution varied the COD loading rate by between
2.7 to 11.5 kg/m
performance and stability and adds valuable nutrients. *
The anaero-b'ic digester for optimum performance must be
i-
fed an equalized waste stream.
XI-5
-------
Recirculating C0? qas (scrubbed to remove H-S) and/or
mechanical agitation of the digester contents improves
the organic matter removal efficiency of the digester
and can lower the retention time necessary for treatment.
« The retention time for anaerobic digestion could be as
short as 5.6 days, although normal performance- requires
about 10 days.
Volatile acid production must be controlled by either
chemical addition or by dilution with other wastewater.
t Gas production is approximately 55 to 65 percent methane
and 45 to 55 percent carbon dioxide.
The effluent from anaerobic digesters treating rum
distillery mostos must have some form of additional
treatment to meet the proposed NPOES limitations.
ANAEROBIC/AEROBIC SYSTEMS
Cosculluela, in Cuba, investigated a biological system for
the treatment of molasses distillery waste (11). His studies
ran for a period of three years. Cosculluel'a recommended that
mostos be diluted four times with water, mixed with a bacterial
culture rich in ammonifying organisms, and then digested in an
open concrete tank for a period of four days. He suggested that
the effluent then be passed through biological filters for addi- ,
tionalstability. ^
Recent research in Russia and Poland has indicated the feasi-
bility of biological systems for the treatment of molasses ferment'
ing and distilling wastes. In Russia, stillage was treated by
XI-6
-------
anaerobic fermentation which reduced the initial BOOT of greater
than 10 g/i by 90 percent. This anaerobic effluent was adjusted
to SODy 800 to 1,300 mg/ii by dilution with municipal wastewater
and treated by activated sludge to give an effluent of 800^ 24- to
60 mg/i, and COD 180 to 240 mg/z (1H) . In Poland, spent molasses
from feed yeast production with a SOOg of 12,870 mg/z and a pH
of 4.4 were anaerobical1y fermented and then treated by the acti-
vated sludge process. Without phosphate nutrient addition the
effluent from the system had a BODg of 70 to 600 mg/z; with phos-
phate addition the effluent BODg concentration was 75 to 300 mg/z
(105).
Bhaskaran (10) reported that laboratory-scale anaerobic digeS'
tion reduced the 800= of molasses distillery slops from about
30,000 mg/4 to the order of 1,000 mg/z. The 800-to-nitrogen ratio
of the digested slops effluent indicated that it was suitable for
activated sludge treatment. In the laboratory, activated sludge
was built u.p by aerating sewage for a number of days. The activa-
ted sludge was then gradually acclimatized to the digester efflu-
ent. When stable conditions of operation were established, the
results showed that aeration for 8 hr effected 93 percent 8005
removal and yielded an effluent with 63 mg/z 3005-
THE ANAMET PROCESS ' "X
>
A proprietary ,jriol ogical system has been developed in Sweden
incorporating anaerobic and aerobic treatment. This process, the
xr-7
-------
Anamet process, differs from other biological systems in that it
has all of the following characteristics:
1. Use of a closed tank for the anaerobic stage rather than
an open pond.
2. Recovery of methane gas for reuse as fuel.
3. Recycle of digester sludge within the anaerobic process
after separation from the digester effluent.
4. Recycle of waste-activated sludge to the digester.
5. Optional nitrogen stripping with recovery of ammonium
salts.
Other biological systems may have some of these characteris-
tics, but Anamet necessarily has all of them.
Applicability to Rum Wastes
The rights to the Anamet process are held by the Swedish
engineering firm of AB Sorigona, located in Staffanstorp, Sweden.
Anamet was developed especially to handle wastes from fermentation
and food industries, rich in carbohydrates and nitrogen compounds.
Anamet has not been used to treat rum distillery wastes, but it
is in full-scale operation at four beet sugar refineries and one
potato alcohol distillery in Sweden. The process has also been
in use for approximately two months on a pilot plant basis at a
beet molasses alcohol distillery in Belgium (.99). According to
representatives of Sorigona, the results of existing Anamet plants
have been excellent, with BOO reductions of up to 99 percent being
achieved. After visiting the facilities of Puerto -ftico Distillers
XI-8
-------
and laboratory testing of the mostos, Sorigona's representatives
expressed the belief that mostos could be treated successfully
with Anamet (49). Subsequently, a pilot plant was constructed.
and began operating in July of 1978. The plant has a rated capa-
city of about 6,000 A/day, or 0.7 percent of the total waste flow.
Initial tests were-begun with a 25-percent-by-volume digester seed
of digested municipal sludge. Preliminary results are not yet
available from this operation.
£quipment and Design
The equipment, used and the design of Anamet plants vary with
different wastewaters and. with the required final BOD (99). Figure
XI-1 presents a typical Anamet system. The anaerobic step is- car-
ried out in a steel tank. The ta.nk has a side-mounted stirrer for
mixing and to keep the sludge suspended. It also has a sedimen-
tation zone- in which sludge is separated from the wastewater and
retained. From the top of the tank, methane gas is drawn off to
a gas supply. The pH and temperature are optimized for the anaero-
bic conversion of the wastes into methane gas.
The aerobic step is carried out as an activated sludge pro-
. cess in a concrete basin or pond. Aeration is performed by sur-
face aerators or in cold climates by submerged aerators.
Sedimentation of aerobic sludge and final clarification of ^
*>
the wastewater are .carried out in a horizontal thickener with rakes
/
M
or in other suitable equipment. Waste sludge is pumped to the
anaerobic tank for stabilization-.
XI-9
-------
ANAEROBIC
DIGESTER
CLARIFIES
EFFLUENT
WASTE
SLUDGE TO
DIGESTER
DIGESTER--
AGITATOR
Figure 'XI-1. Typical Anamet System
XI-10
-------
Where nitrogen reduction is involved, it is achieved by strip
ping the water with air after the anaerobic step. Most of the
organic nitrogen compounds have by then been decomposed and the
nitrogen remains as ammonium ions. Lime is added in order to
increase the pH of the solution and this frees the ammonia before
stripping. The ammonia is absorbed and a concentrated ammonium
salt solution is produced which permits recovery of ammonia for
reuse as a. nutrient in the fermentation process.
Process Performance
Tha specific improvements in performance which Anamet systems
attempt to provide over other anaerobic/aerobic systems include:
Low overall sludge generation
t Short digester retention time
* High methane gas production
Excellent 800 removal.
These features are discussed below.
In the Anamet process the sludge created in the aerobic stage
is recycled into the anaerobic stage and converted into methane
and carbon dioxide, thus reducing to a minimum the amount of
solid waste which would have to be disposed of upon completion of
the process. A Swedish sugar factory using the Anamet process ;
reported no surplus sludge generated for the first three months '\.
"*»
after start-up of the. full-scale plant (99).
/
The hydraulic retention time for the wastewaters in the anaer
obic step may vary between two and 15 days, depending upon the
nature of the wastewater. In order to avoid extensive digester
XI-11
-------
retention times, which would be a problem in the treatment of rum
distillery wastes, the anaerobic sludge is retained in the system.
This is usually achieved by separation in a settler and recircula-
tion of the thickened sludge. The accelerated activity in the
anaerobic step is also facilitated by the return of the biosludge
from the aerobic step. This sludge acts as a "support" for the
enzymes of the anaerobic bacteria (99).
The operating conditions of the system are maintained so as
to maximize the anaerobic conversion of the waste into methane gas
The amount of gas produced depends on the wastewater and is repor-
ted to vary from 0.3 to 0.5 m CH./kg BOD, digested. Sorigona
suggests that the digester operating parameters as shown in
Table XI-1 be maintained to optimize the conditions for the
methane-forming bacteria (49).
TABLE XT-! . .RE.CQMMEMDEQ' OPERATING. PARAMETERS
FOR ANAEROBIC DIGESTION IN THE ANAMET SYSTEM (49)
Parameter Typical Operating Ranoe
pH 6.8 - 7.5
Temperature 35 - 60°C
Bacterial retention time >6 days
Wastewater retention time 2-20 days
Process toxicity levels NH. <1,000 mg/i
r "
K+, Na + < 5,000 mg/z
H2S < 100 mg/L
XI-12
-------
Wastewaters with high inorganic salt concentrations, such as
mostos, may tend to increase the minimum bacterial retention time
shown in Table XI-1. In addition, there, are toxic effects from
high concentrations of inorganic compounds such, as ammonium, potas-
sium, sodiuum, and hydrogen sulfide in the wastewater. By adapt-
ing the bacteria, however, it may be possible to operate at con-
centrations higher- than the normal toxic levels shown in Table
XI-1 C49],
BOD reductions with Anamet vary depending upon the specific
wastewater and. the retention time, but Sorigona claims that BOD
reductions of up to 99 percent have been experienced in the treat-
ment of wastes other than rum distillery effluents. Sorigona
reports that in various situations when the Anamet process has
been used, between 70 and 90 percent of the original 800 was
removed in the anaerobic step, leaving 10 to 30 percent to be
treated in the aerob.ic step. This high degree of 800 reduction
in the anaerobic step is primarily attributable to the production
of methane gas. Methane has a low water solubility and, conse-
quently, nearly all of it escapes from solution, thereby signifi-
cantly reducing BOO (99).
The BOO reduction which has been reported for the second step
of the Anamet process is very high, with up to approximately 93 >
'percent reduction reported for some wastes. The efficiency of the
300 reduction in tire aerobic step is attributed to the anaerobic
pretreatment where the large molecules are hydrolyzed into sub-
stances which are easily assimilated. Moreover, a sufficient
XI-13
-------
amount of inorganic nitrogen remains in the water to facilitate
the conversion of the biomass in the aerobic step (99),
A complete list of Anamet facilities existing and under con-
struction is contained in Appendix B. it includes the type of
waste treated and the BOD5 reduction achieved at each plant, as
reported by Sorigona (50). The types of wastes which have been
treated with Anamet generally have had BOOr and COO concentrations
considerably lower than those of Caribbean mostos (49,95,99). BOOg
concentrations have been in the range of 4,000 to 13,000 mg/2, for
many of the wastes treated, while mostos has a typical BOOg concen-
tration of about 35,000 mg/i. Therefore the results which will
become available from Puerto Rico Distillers' pilot Anamet facility
at Arecibo are of great importance in determining the adaptability
of the process to mostos.
Costs
The pilot plant results are also critical to determing the
costs of a full-scale operation. According to Sorigona (99), the
investment costs for an Anamet plant-are generally low, on the
order of $50 to $100/kg of BOD, removed per day, due to the compact-
ness of the plant. Operating costs are reportedly negligible since
no nutrients are needed, and energy as well as personnel require-
ments are low. Sludge handling costs are also small. The value'..
of methane gas produced is said to more than cover the operating
i"
costs for many applications (99).
XI-14
-------
A8 Sorigona has provided Puerto Rico Distillers with a turn-
key investment price estimate for a full-scale facility (ad). The
estimate is for a loading rate of 65,000 Ib of 800/day, 225,000 gal
of mostos/day, and assuming' 95 percent 3005 reduction. The turnkey
investment cost estimate submitted April 4, 1978, in U.S. dollars
was $2.55 x 10 . This is -a cost of approximately $91/kg of BOO-
removed per day. Methane gas production was estimated at the equi-
valent of Z,340 tons of fuel oil per year. Electrical power con-
sumption was estimated at 1 x 10 kWh/yr; and sludge, production
at 6,700 m /yr. The cost of sludge transport was calculated at
$57ton, but this does not include land (EQ).
One. of the relative disadvantages of any of the anaerobic/
aerobic biological treatment systems is that the effluent remains
very dark in color. It would be technologically possible to
remove the color from the effluent by means of large quantities
of activated carbon or by ozone oxidation. Either method would
be prohibitively expensive, however (24).
XI-15
-------
CULTIVATION OF FODDER YEASTS, BACTERIA, AND FUNGI IN SLOPS
The propagation of single strains of various yeasts, bacteria,
and fungi in waste liquids is sometimes referred to as single-cell
protein synthesis. Since these organisms can metabolize the organic
materials of a substrate which itself conta.ins relatively little
protein, the high protein content of the biomass represents a trans-
formation of the nitrogen of the substrate to form a protein bio-
synthesis. Previous investigations utilizing rum slops as a sub-
strate have been performed primarily with certain yeasts and molds
(considered fungi). When extracted and dried, the yeast or mold
biomass (containing approximatley 40 to 46 percent protein) has a
relatively high market value as an animal feed supplement (52).
A flow sheet for yeast manufacturing from molasses alcohol
stillage, as it was performed at Hsinying By-Product Factory (Zl),
is shown in Figure XI-Z,. Preceding the fermentation of the stillage,
yeast inoculum was prepared using a shaking cultivation and two
seed tanks. Molasses was employed for the culture to maintain
strong and active yeast. The stillage coming from the distillery
was cooled, but neither dilution nor sterilization was required.
Stillage was continuously fed to a fermentation tank by displace-
ment. Supplemental nitrogen and phosphorus were provided by the
addition of a cream solution of urea, ammonium sulfate, and calcium
superphosphate. Airr was provided with a turbo blower. The tempera-
ture was regulate'cl by a refrigeration cooling system designed with
a heat exchanger to eliminate the possibility of contamination of
the fermenter. The fermented stillage flowed from the tank to a
XI -1 6
-------
YEAST
SEED
SHAKING
CULTIVATION
WASTE
LIQUOR
x
ii
i
i
STILLAGE
NUTRIENT
SOLUTION
FEED TANK
COOLING
TANK
SEPARATORS
FERMENTATION
TANK
AIR
BLOWER
HEAT
EXCHANGE
TURBO
COMPRESSOR
REFRIGERATING
MACHINE
HAMMER
MILL
DRIED
YEAST
PRODUCT
HEAT
EXHAUST
REF.
Figure XI-2, Yeast Cultivation in Molasses Distillery Stillage*
-------
yeast separator. The separated yeast went through a drum dryer
and hammer mill to form the final product.
Yeast cultivation of molasses alcohol stillage has been inves-
tigated in Taiwan (21, 64). Successful work has been performed
with Torula uti1is, which has a protein content of 46 percent. One
ton of dry torula yeast was produced from 60 to 65 tons of stillage
in a small, factory-scale aerobic fermenter. In spite of these
excellent results, it is expected that in most cases up to 100 tons
or more of stillage may be required to yield one ton of dry torula
feed yeast (47), mainfy due to the high BODr of the Taiwanese
stillag-e (approximately 60,000 ppm) . Stillage BODr values from
Puerto- Rican rum processors, in contrast, have been shown to be
lower (average total BOD values were reported to be 32,900 and
54,900 ppm for 6 and 5 samples, respectively) (93).
Researchers in Taiwan (21) have, reported that the carbon source
for yeast growth is derived principally from the organic matter por-
tion of the stillage. In laboratory and factory fermentations, the
pH increased rapidly from 6.8 to 7.9 during yeast formation, indi-
cating that much of the energy being used was from volatile acids,
lactic acid, and other combined organic acids in the stillage. Stil-
lages from beet and cane molasses are different in organic composi-
tion: the relative ash content of cane molasses is about 40 percent
higher than that of,-its beet counterpart (96). It is well known
that beet molasses provides a better substrate for yeast production
than cane molasses. Therefore, it is expected that beet molasses
stillage will make a better substrate than cane molasses stillage.
XI-18
-------
In the Taiwanese experiments (64) with Torn la utilis, the BOOg
concentration of the stillage was reduced by 33 percent, and the
total solids concentration by 29 percent. In Soviet experiments
(51) with de-yeasted beet molasses stillage, the cultivation of
various fodder yeasts decreased the 8005 by 41 to 54 percent, and
the total solids by approximately 37 percent. The Soviet experiments
involved dilution of the stillage prior to yeast cultivation, as
have experiments carried out by French investigators (61). These
latter experiments indicated that dilution promoted the yeast
activity.
The amount of income to be derived from yeast production is
dependent upon the yeast yield and its marketability. The produc-
tion costs may vary considerably, depending upon the extent to which
additional nutrients must be supplied for the process, the retention
times required, and the steps necessary to separate the yeast from
the liquid. Research in some of these areas (30) is underway at
Puerto Rico Distillers' laboratories in Arecibo, Puerto Rico. Exper-
iments have already been conducted using Aspergi11 us phoenici us
(a fungus) in fermentation tanks with up to 1,500-liter volumes.
The highest SOOg: reduction achieved in these tests was 59 percent
at a retention time of seven days. Puerto Rico Distillers (80)
regards a retention time of greater than two to three days as exces-
sive to be useful in-a full-scale operation, due to a lack of
^*
space and the high costs for tanks. These experiments also yielded
disappointing results concerning the separation of the yeast from
solution. 3ased on earlier experiments of a smaller seals at the
-------
Rum Pilot Plant of the University of Puerto Rico, Puerto Rico
Rum Distillers had hoped for strong formation of a mycelium, a
visible ma'ss of microbial cell matter, which would easily be
removed from solution. However, mycelium formation in the experi-
ments was not as consistent as had been expected. Further infor-
mation in these areas, and on supplementary nutrient requirements
as well, is needed to make .production cost estimates.
Another important cost consideration is electrical power. Power
is consumed primarily for aeration purposes, as many as 400 volumes
of air being required per volume 'of rum slops (61). Power require-
ments for the operation of a cooling system may also be considerable
since a large amount of heat is evolved during feed yeast cultiva-
tion, and must be continuously removed to safeguard fermentation
activity and maximize yield.
After yeast cultivation and separation, the slops will still
pose a substantial treatment problem due to the remaining high B005
level. Because much of the organic matter of the waste will have
been used in producing yeast, the value of the waste in the pro-
duction of a gas by anaerobic digestion will be reduced. Its min-
eral content would be extremely high if evaporated to CMS for use
as an animal feedstuff and, if the CMS were incinerated, the amount
of heat recoverable for the needs of evaporation would be lower.'
On the positive side-, however, the reduced strength of the waste
could improve the operation and performance of aerobic biological
treatment.
XI-20
-------
( REVISED )
XII. BIBLIOGRAPHY
r.
Allen, D.M., R.L. Sargent, and J.A. Obertreuffer. Treatment
of combined sewer overflows by-high gradient separation.
EPA/600/2-77/015, Sala Magnetics, Inc., Cambridge, Massachu-
setts, March 1977. 130p.~ (Available from National Technical
Information Service (NTIS) as P8-264 935).
Ammerman, C.8., P.R. Henry, S.G. Potter,
Effect of condensed molasses solubles as
for cattle. Presented at the Xlth International
Nutrition, Rio Oe Janeiro, August 1978.
and H.M. Seeker.
a feed ingredient
Congress
of
APV Company, Inc. APV processing equipment.
Tonawanda, New York, 1978. 19p.
Bulletin CP-475
4. Bacardi Corporation. Written testimony presented at .adjudi-
catory hearings for Virgin Islands Rum Industries, Limited,
Puerto Rico Distillers and Schenley Industries, Inc., 1978."
4a.
5.
Backus, A.A.
1 ery slops.
and C. Haner, Jr. Volatile acids from distil-
U.S.P. 1,396,006 and 1,500,131, November 8, 1922
Basora and Rodriguez, Consulting Engineers. Facilities plan-
ning for the Bayamon W.astewater Region. Draft environmental
impact statement. Prepared for Puerto Rico Aqueduct and
Sewer Authority, Commonwealth of Puerto Rico, March 1975.
5. Basora and Rodriguez, Consulting Engineers. Facilities plan-
ning for the Bayamon Wastewater Region. Vol. II. Engineering
report. Prepared for Puerto Rico Aqueduct and Sewer Author-
ity, Commonwealth of Puerto Rico, June 1976.
7. Basora and Rodriguez, Consulting Engineers. Facilities plan-
ning for the Bayamon Wastewater Region. Vol. IV. Oceanogra-'
phic study. Prepared for Puerto Rico Aqueduct and Sewer
Authority, Commonwealth of Puerto Rico, March 1976.
8. Basu, A.K. Characteristics of distillery wastewater. J.
Water Pollut. Control Fed., 47:2184-2190, 1975.
XII-1
-------
9. Basu, A.K., and L. leclerc. Comparative studies on treat-
ment of beet molasses distillery waste by thermophilic and
mesophilic digestion. Water Res., 4:103-104, 1975".
10. Bhaskaran, T.R. Utilization of materials derived from treat-
ment of wastes from molasses distilleries. Adv. Water Pol-
lut. Res. , 2(2):85-104, 1964.
11. Biaggi, N. Studies on rum distillery wastes in Puerto Rico.
Ph.D. dissertation. University of Michigan, Ann Arbor, 1956.
12. Bolton, D.H., O.A. Hines, and J.P. Bouchard. The applica-
tion of the ICI Deep-Shaft Process to industrial effluents.
Proc. Ind. Waste Conf., 31:344-351, 1976.
13. Boruff, C.S. Recovery of fermentation residues as feeds.
Ind. Eng. Chem. , 39:602-607, 1947.
14. Bradley, B.J. ICI Deep-Shaft Effluent Treatment Process.
Presented at the 24th Annual Meeting Canada Section, Ameri-
can Water Works Association, Halifax, Nova Scotia, September
12-15, 1976.
15. Srau, H.M., Puerto Rico Distilleries, Inc. Supplemental
testimony before the United States Environmental Protection
Agency in the matter of National Pollutant Discharge Elimir_
n.ation System Permits for Puerto Rico Distillers, Inc.,.
Schenley Industries, Inc., and Virgin Islands Rum Industries,
Ltd.
16. Brooks, N.H. Conceptual design of submarine outfall. I.
Jet diffusion. Technical Memorandum 70-1, California Insti-
tute of Technology, Pasadena, W.M. Keck Laboratory, 1970.
19p.
17. Brooks, N.H. Conceptual design of submarine outfall. II.
Hydraulic design. Technical Memorandum 70-2, California
Institute of Technology, Pasadena, W.M. Keck Laboratory,
1970. 12p.
18. Brooks, N.H. Dispersion in hydrologic and coastal environ-
ments. Report No. KH-R-29, California Institute of Techno-;
logy, Pasadena, December 1972. 136p.
19. Burnett, W. F.. '-Rum distillery wastes: laboratory studies on
aerobic treatment. Water Sewage Works, 120:107-111, 1973.
XII-2
-------
20. Chakrabarty, R.N. Potash recovery - a method of disposal of
distillery wastes and saving foreign exchange. Presented at
Symposium in New Developments of Chemical Industries Relat-
ing to Ethyl Alcohol, Its By-Products and Wastes, New Delhi,
October 14-16, 1963.
ZT . Chang, C.T., and W.I. Yang.. Study of feed yeast production
from molasses distillery still age.. Taiwan Sugar, 20:200-205,
1973.
22. Chatterjee, A.C. Production of fertilizer from distillery
spent wastes. Presented at the 38th S.T.A. of India, Kan-
pur, 1972..
23. Conklin, O.E., Bodega Marine Laboratory, University Of Cal-
ifornia. Personal communication to M.A. Caponigro, Septem-
ber 22, 1978.
24. Crowley, 8.A., Schenley Distilleries, Inc. Supplemental
testimony before the United States Environmental Protection
Agency for the matter of National Pollutant Discharge Elimi-
nation Permits for Puerto- Rico Distillers, Inc., Schenley
Industries, Inc., Virgin Islands Rum Industries, Ltd.
25. Cruver, J.E., and J. Nusbaum. Application of reverse osmo-
sis to wastewater treatment. J. Water Pollut. Control Fe.dT",
46:301-311, 1974.
2'6. Gulp, R.L., and G.L. Gulp. Advanced Wastewater Treatment.
Van Nostrand Reinhold, New York, 1971. 3l8p.
27. Damron, 3.L. Condensed molasses solubles in poultry feed;
final research summary to George H. Oorion, Bacardi Corp.,
Jacksonville, Florida.
28. Oedert Corporation. Dependable process equipment. Chicago
Heights, Illinois, 1978. 9p.
29. Diaz-Medina, M. and P.F. Randel . Dried rum distillery s'til-
lage in laying rations. J. Agric. Univ. P.R., 62:149-155,
1978;
30. Dorian, G., Bacardi Corporation, Personal communication to *
K.V. LaConde, July 18, 1978.
j-
31. Dorian, G., Bacardi Corporation. Personal communication to
K.V. LaConde and C.J. Schmidt, July 20, 1978.
32. Dorian, G., Bacardi Corporation. Personal communication to
K.V. LaConde, September 1, 1978.
XII-3
-------
33. Dorian G., Bacardi Corporation. Personal communication to
K.V. LaConde, September 28, 1978.
34. Dorian G., Bacardi Corporation. Submitted testimony before
the United States Environmental Protection Agency in the
matter of National Pollutant Discharge Elimination Systems
Permits for Puerto Rico Distillers, Inc., Schenley Indus-
tries, Inc., and Virgin Islands Rum Industries, Ltd.
35. Edwards, J.Y., Pacific Molasses Company. Supplemental tes-
timony before the Environmental Protection Agency in the
matter of National Pollutant Discharge Elimination System
Permits for Puerto Rico Distillers, Inc., Schenley Indus-
tries, Inc., and Virgin Islands Rum Industries, Ltd.
36. Environmental Science and Engineering. Draft development
document for effluent limitations and new source performance
'standards: miscel 1 arreous foods and beverages point source
category. Gainesville, Florida, March 1975. 5 vols.
37. Falcman, W. Combustion of molasses slops as a solution of
the wastewater 'problem in the alcohol industry. Szeszipar,
19:103-105, 1971.
38. Gabel , D.H., Lederle Laboratories. Personal communication,
March 1978.
3-9. Gartner, R.J.W, and P.K. O'Rourke. Effects of antibiotics,
.dried molasses distillers solubles and zeranol in all-sorg-
hum srain rations fed to steers. Aust. J. Exp. Agric. Anim.
Husb. , 17:214-220, 1977.
40. Gillies, M.T. Animal Feeds from Waste Materials. Noyes
Data Corporation, Park Ridge, New Jersey, 1978. 346p.
41. Gross, R.W., Jr. High purity oxygen activated sludge pro-
cess for industrial wastewater. Union Carbide Corporation,
South Charleston,-West Virginia.
42. Gyger, R.F., and E.L. Doerflein. Save energy - use 02 on
wastes. Hydrocarbon Process., 55:96-100, 1976.
43. Haner C. U.5.P. 1,430,160.
44. Harada, Y., K.'"Hirano, and T. Koike. Closed system in fer-
mentation was'tewater treatment. Presented at Chemical Engi-
neering Congress, Denver, August 29, 1977. 5p.
XII-4
-------
45
46
47
48
49
50
51
Hartner, M., Savannah State College. Personal communication
to M.A. Caponigro, October 9, 1973.
Heres Gonzalez, A. Personal communication to K.V. LaConde,
September 22, 1978.
Heres Gonzalez, A. Personal communication to K.V. LaConde,
October 13, 1978.
Hiatt, W.C. :
tion of rum
28:966-976,
A.O. Carr, and J.F
disti11ery wastes .
1973.
Andrews.
Proc. Ind.
Anaerobic diges-
Waste Conf.,
Huss, L. Methane production from Cassava Distillery stillage
with the Sorigona WWTS. Presented at the Semario Interna-
cional Sobre Tratamento de Vinhoto, Rio Oe Janeiro, August
1976. 6p. -
Huss, L., A.8. Sorigona. Personal communication. April 1978
Il'ina, L.O., N.G. Sitnik, E.I. Podgurskaya, M.S. Pavlovskaya
and L.V. Rudnichenko. An investigation of the chemical com-
position of ololasses stillage with an evaluation of the
degree
30-34.
of its purification. Fermentn. Spirt. Prom.,. 1974
52
53
54
55
56
Jackson, C.J. The treatment of distillery and antibiotics
wastes. . In: Waste Treatment; Proceedings of the Second Sym-
posium on the Treatment of Wastewaters, 1959. Pergamon
Press, New York, 1960. pp. 226-239.
Kawano , T.
and method
fi1ed June
, H.
of
27,
Kojima, H
aqui animals
Ohosawa, and K. Morinaga. Feed
cultivation. Patent 3",939,279,
1973, awarded February 17, 1976
Knezek, B.D., and R.H. Miller, eds. Application of sludges
and wastewaters on agricultural land; a planning and educa-
tional guide. North Central Regional Research Publication
235, Ohio Agricultural Research'and Development Center, Woo-
ster, October 1976. 98p.
Koh, R.C.Y., California
Personal Communication,
Institute of Technology, Pasadena.
July 1978.
Koh, R.C.Y. , a-nd N.H. Brooks
disposal in tfhe ocean. Annu
1975.
Fluid mechanics of
Rev. Fluid Mech., 7
wastewater
187-211 ,
XII-5
-------
57. Krabbe, E., U.S. EPA Region II Facilities Technology Divi-
sion. Testimony in rebuttal to Bacardi Corporation for the
adjudicatory hearing in the matter of Puerto Rico Distillers,
Inc., Schenley Industries, Inc., and Virgin Islands Rum
Industries, Ltd.
58. Krabbe, E., U.S. EPA Region II Facilities Technology Divi-
sion. Testimony in rebuttal to Schenley Industries, Inc.,
and Virgin Islands Rum Industries, Ltd., for the adjudicatory
hearing in the matter of Puerto Rico Distillers, Inc., Schen-
ley Industries, Inc., and Virgin Islands Rum Industries, Ltd.
5.9. Krabbe, E., U.S. EPA Region II Facilities Technology Divi-
sion. Testimony in rebuttal to Puerto Rico Distillers, Inc.
for the adjudicatory hearing in the matter of Puerto Rico
Distillers, Inc., Schenley Industries, Inc., and Virgin
Islands Rum Industries', Ltd.
60. Krabbe, E., U.S. EPA Region II Facilities Technology Divi-
sion. Testimony of Erik Krabbe for the adjudicatory hearing
in the matter of Puerto Rico Distillers, Inc., Schenley
Industries, Inc., and Virgin Islands Rum Industries, Ltd.
61. Kujala, P., R. Hull, F. Engstrom, and E. Jackman. Alcohol
from molasses as a possible fuel, and the economics of dis-
tillery effluent treatment. Sugar Azucar, 1976:28-39.
62. Kulperyer, R.J., and L.C. Matsch. Comparison of treatment
of problem wastewaters with air and high purity oxygen .acti-
vated sludge systems. Presented at the International Asso-
ciation of Water Pollution Research on Design - Operation
Interactions at Large Wastewater Treatment Plants, Vienna,
September 8-12, 1975.
63. LaConde, K.V., R.J. Lofy, and R.P. Stearns. Municipal sludge
agricultural utilization practices - an environmental assess-
ment. EPA Contract 68-01-3265, SCS Engineers, Long Beach,
California. (In press.)
64. Lai, C.L., Y.T. Chuang, and S.L. Sang. A research report on
the treatment of agriculture and industrial waste. Taiwan
Sugar, 21 :37-41 , 1974.
65. Lesoffre et Cie. Substituting ammonium for potassium in by--'
products from -the sugar and fermentation Industries. French
Patent 1,497^.480, filed March 29, 1966, awarded October 13,
1967.
66. Lopez, J.I., Puerto Rico Distillers. Personal communication,
September 29, 1978.
XII-6
-------
67. Lopez, J.I., Puerto Rico Distillers. Personal communication,
October 4, 1978.
68. Lyman, C.M., K.A. Kuikan, and F. Hale. Essential amino-acid
content of farm feeds. J. Agr. Food Chem. , 4:1008-1113,
1956.
59. Maphis, S.W., Briscoe Maphis. Personal communication to K.
V. LaConde, October 10, 1978.
70. Mattson, M.E. Membrane desalting gets big push. Water
Wastes Eng., 12:35-42, April 1975.
71.- Nelthropp, C., Virgin Islands -Rum Industries, Limited. Per-
sonal communication to K.V. LaConde and C.J. Schmidt, July
2.7, 1978.
72. Olszewski, C. Separation of potassium Compounds in'view of
their reuse. French Patent 1,377,487, filed September 24,
1963, awarded November 6, 1964.
73. Oral testimony presented at adjudicatory hearings for Virgin
Islands Rum Industries, Inc., Puerto Rico Distillers, and
Old San Juan, July 1977.
74. Page, A.L., Department of Soil and Environmental Science,
University of California, Riverside. Personal communication
to K.V. LaConde, September 26, 1978.
75. Perez Escolar, R. Reclamation of a saline-sodic soil fay use
of molasses and distillery Slops. J. Agric. Univ. P.R.,
50:209-217, 1966.
7'6. Perez Escolar, R. Separation and characterization of the
active soil aggregating agent present in distillery slops.
J. Agric. Univ. P.R., 51:304-308, 1967.
77. Perez Escolar, R. Stability of soil aggregates treated with
distillery slops or blackstrap molasses. J. Agric. Univ.
P.R., 50:174-185, 1966.
78. Perez Escolar, R. Use of molasses and distillery slops with
sulfur for the reclamation of a saline-sodic soil from Puerto
Rico. J. Agric. Univ. P.R., 51:55-65, 1967.
*
79. Phung, T., Lv Barker, 0. Ross, and D. Bauer. Land cultiva-
tion of industrial wastes and municipal solid Wastes; state-
of-the-art study. EPA/600/2-78-140a , SCS'Engineers , Long
Beach, California, August 1978. 2 vols.
X I I - 7
-------
80. Puerto Rico Distillers, Inc. Proposals for the treatment of
rum distillery wastes. Submitted with Schenley Industries,
Inc., and Virgin Islands Rum Industries, Ltd.
81. Radhakrishnan, I., S.B. Oe, and B. Nath. Evaluation of the
loading parameters for anaerobic digestion of cane molasses
distillery wastes. J. Water Pollut. Control Fed., 41:R431-
R440, 1969.
82. Ramirez Rivera, R., and H.M. Brau, Puerto Rico Distillers.
Personal communication to K.V. LaConde and C.J. Schmidt,
July 21, 1978.
83. Reich, G.T. Production of carbon and potash from molasses
distillery. Trans. Am. Inst. Chem. Eng. , 61:233-252, 1945.
84. Rivera, L.H. Soil survey, Virgin Islands of the United
States. U.S. Government Printing Office, Washington, D.C.,
1970. 81p.
85. Roberts, N'., and J.B. Hardwick. High rate trickling filter
pilot plant studies in distillery Waste. Proc. Ind. Waste
Conf., 6:80-87, 1951.
86. Roberts, R.C Soil survey, Puerto Rico. Soil Survey Report
1936 Series, No. 8, U.S. Government Printing Office, Washing-
ton, D.C., 1942. 503p.
87. Rodriguez, J., Bacardi Corporation. Supplemental testimony
to the United States Environmental Protection Agency in the
matter of National Pollutant Discharge Elimination System
Permits for Puerto Rico Distillers, Inc., Schenley Industries,
Inc., and Virgin Islands Rum Industries, Ltd.
88. Rossoff,- G., Standard Brands. Personal communication to K.
V. LaConde, August 17, 1978.
89. Rubin, A.J. Removal of trace metals by foam separation pro-
cesses. J. Am. Water Works Assoc., 60:832-846, 1968.
90. Sanford, D., Eco-Research Limited. Personal communication,
August 1978.
91. Sawyer, C.N., and E.J. Anderson. Aerobic treatment of rum
wastes. Water-Sewage Works, 96:112-114, 1949.
^
92. Sen, B.P., and T . R. Bhaskaran. Anaerobic digestion of liquid
molasses distillery wastes. J. Water- Pollut. Control Fed.,
34:1015-1025, 1962.
XII-8
-------
93.
94.
96.
97.
Shea, T.G., E. Ramos, J. Rodriguez- and G.H. Oorion. Rum
distillery slops treatment by anaerobic contact process.
Bacardi Corporation, San Diego, Puerto Rico, July 1974. 109p
(Available from National Technical Information Service (NTIS)
as P8-238 291 ) .
Ski rstymonski i , A.I
tial elimination of
tion. Perm. Spirt.
, M.S. Sushi, and T.N. Pukhovaya. Par-
salt from vinasse during its concentra-
Prom., 37(7):26-27, 1971. (Russian).
95. Skogman, H.
tion wastes
Bi ochemi cal
Effluent treatment of molasses oased fermenta-
Presented at the Effluent Treatment in the
Industries Conference, October 1977.
Soldevila, M., and M. Rojas Daporta.
levels of sugar cane molasses on egg
Univ. P.R., 60:531-634, 1976.
fferent
Effect of di
production. J. Agric
Soldevila, M., and R. Irizary. Influence of dry rum distil
lery slops in diets for white leghorn hens on laying rate,
feed conversion, and egg size and quality. J. Agric. Univ.
P.R., 61:465-469, 1977.
98.
99.
100.
Sopper, W.E., and
Treated Muni ci pal
E.PA/560/2-74-003,
Washington, O.C.,
L.T. Kardos, eds.
Wastewater through
U.S. Environmental
1974. 463p.
Conference
Forest and
Protecti on
on Recycli ng
Cropland
Agency ,
Sori gona,
bury, New
A..B . The Amamet
York, 1978. 8p.
Process. Chemapec, Inc., Wood-
Sowa, W. Trip report for Association of Canadian Distillers'
Third Liquid Feed Symposium, American Feed Manufacturers
Association. Report ORF-73-2, Ontario Research Foundation,
Ottawa, Canada, 1973.
101. Stern, A.M., L.L. Gasner, and F.A. Sanders. An economical
low-energy bio-oxidation process for the treatment of high
BOO wastewaters, Proc. Ind. Waste Conf., 29:881-888, 1974.
102. Stros , F., R. Petkvova, V., and V. Varvazovsky. Desalted
molasses stillage as cattle feed. Kvasny Prum., 13:155-157,
1967. (Czech). '-,
103. Struzeski, E.,,-EPA. Personal communication to K.V. LaConde,
October 12, 1^78.
104. Struzeski, E., EPA. Personal communication to H. Thron ,
October 17, 1978.
X11 - 9
-------
105. Tolaney, M. Treatment of high strength citrus wastewater
with high purity oxygen activated sludge process. Proc.
Ind. Waste Conf., 30:171-183, 1975.
106. Tomczynska, J. Activated sludge as a second stage of bio-
logical treatment of wastewaters from molasses slops-process
feed yeast production. Pr. Inst. Lab. Badaw. Przem. Spouzyw.
24:75-93, 1974.
107. Tosteson, T.R., B.R. Zaidi, D. Hale, and K. Verner. The
effect of the mosto on the growth of marine microorganisms.
Report No. RPP:l-73, Agricultural Experiment Station, Uni-
versity of Puerto Rico, Mayaguez, October 1973. 35p.
108. U.S. Crop Reporting Board. Commercial fertilizers: consump-
tion in the United States. Sp. :Cr. 7, U.S. Department of
Agriculture, Washington, O.C., Statistical Reporting Ser-
vice, May 1971 .
109. U.S. Crop Reporting Board. Commercial fertilizers: consump-
tion of commercial fertilizers, primary plant nutrients and
micronutrients . Statistical Bulletin 472, U.S. Department
of Agriculture, Washington, D.C., November 1976. 64p.
110. U.S. Environmental Protection Agency. Environmental Protec-,
tion Agency ocean dumping regulations and criteria (40 CFR
220-229... .43 FR 1071, January 6, 1978). Environ. Reporter:
Fed. Regul., 131:8101-8133.
111. U.S. Environmental Protection Agency. Effluent Guidel.ine-s
Division. Development document for interim final and pro-
posed effluent limitations guidelines and new source perfor-
mance standards for the fruits, vegetables and specialties
segments of the canned and preserved fruits and vegetables
point source category. EPA 440/1-75/046, Group I, Phase
II. October 1975. 520p.
112. U.S. Federal Water Pollution Control Administration.
Advanced waste treatment research program, July 1964-July.
1967; 'summary report. Cincinnati, Ohio, 1968. 103p.
(Available from National Technical Information Service (NTIS)
as PB-216 647).
113. Virgin Islands Sewage Treatment Plant. Personal communica-"
tion to K.V. kaConde, October 13, 1978.
114. Vitkovskaya, V.A., Y.M. Kravets, E.S. Adamenko, M.D. Lagu-
tenko, T.I. Shamatko, L.R. Zharova, and Y.A. Karanoy. Bio-
chemical purification of wastewaters of .alcohol plants pro-
cessing molasses. Biokhim. Ochistka Stohnykh. Vod. Mater.
Resp. Konf. , 1974:25-26.
XII-10
-------
115. Wallach, A., and A. Wo 1 man. Treatment of distillery wastes
Sewage Works J., 14:382-401, 1942.
116. Wechsler, I., Sala Magnetics, Inc. Personal communication
to M. Montgomery, September 26, 1978.
117. Whitaker, M.C. U.S.P. 1,504,730.
118. Zerban, F.W., and L. Sattler. Unfermentable reducing sub-
stances in molasses. Identification of d_-Allulose. Ind.
Eng. Chem., 34:1180-1188.
X11 -11
-------
APPENDIX A
Aquaculture Correspondence
A-l
-------
UNIVERSITY Ol1' CAI.II-'OHNIA
HCKKKLKY IMViS IKVtNK I .OS AWiKI.JvS KIVERSIOI-: .fAN WK<:fl .SA.V KIIANCI.SCO l|'KBF-JPv«K]oll SANTA IIAftllAllA .SANTA C.WV.
'
AQUACULTUflli PROGRAM ' <> »ox -1'
BODEGA MARINE LABORATORY DOIJECA DAY. CALIFORNIA S)4923
(7071 875-3662
September 22, 1973
Michael A. Caponigro
SCS Engineers
4014 Long Beach Boulevard
Long Beach, CA 90807
Dear Mr. Caponigro:
I've examined the material you sent concerning the waste characteris-
tics of molasses slops. My initial judgment is that this material is not
readily adaptable to use for crustacean aquaculture in terms of a feed
ingredient. It may have some usefulness in a more complex aquaculture
scheme involving its use as a nutrient source for algae or some sort of
detrital feeder which in turn could be utilized by higher organisms.
Along these lines, I have forwarded the material to Dr. Michael Hartman,
Professor, Savannah State College, for his input. Dr. Hartman who has
had extensive experience with a wide variety of aquaculture systems, both
here and at the University of Delaware, is probably the best suited to
evaluate any possible utilization of this product in relationship to
aquaculture.
If I can be of further assistance on matters relating to crustacean
aquaculture in the future, please don't hesitate to write.
Sincerely,
/ N
«>
Douglass. ConJclin
Associate Director
DEC: ta
cc: M.C. Hartman
A-2
-------
SAVANNAH STATE COLLEGE
STATE COCHOt aflAMGH
SAVANNAH, QEOROU
DIVISION or October 9, 1978
NATURAL 3CIINCXS
Mr. Michael A.. Caponigro
SCS Engineers
4014 Long Beach Blvd.
Long Beach, California 90807'
Dear Mr. Caponigro:
Dr. Conklin, a past associate of mine, forwarded your material to me.
After reading over it carefully, I'm afarid that there would not be too"
much use for the waste molasses slops in an aquaculture foodchains. My
evaluation is based on several factors: The 800 and COO are very high,
(table 4, p.44) indicating that the material is very high in organic
carbon compounds. This is further substaniated by the high carbohydrate
content, menti oned. in a Tatar part'of the report. These compunds being
very water souble enhance the growth of Hetertrophic bacteria which,
besides creating a situation of lower dissolved oxygen, they can destroy the
bacterial flora of a biological filter system should such a system be in use.
Table 4 (p. 45) gives a Mg/L figure for copper as ranging from 2.0-124
with 32.3 as the mean. A concentration this high would be toxic to most
shellfish, and acutely toxic to the larval stages. To introduce this
material into an aquaculture foodchain, should it be taken up, would further
concentrate the heavy metals.
Sincerely,
Michael Hartnan, Ph.D.
Assistance Professor
Department of Biology
MH/ls
A-3
-------
APPENDIX 3
Anamet Plants
8-1
-------
Appendix B
REFERENCE LIST
Anamet plants under operation:
Swedish Sugar Company
Ortofta Factory
24100 ESLOV - SWEDEN
Capacity
Reduction
Construction year
Attention: Plant Manager Per Schmidt
Swedish Sugar Company
Hasslarps Factory
Capacity
Reduction
Construction year
Swedish Sugar Company
Karpalund Factory
Capacity
Reduction
Construction year
Swedish Sugar Company
Jordberga Factory
Capacity
Reduction
Construction year
20 ton BOD5/day
98% BOD5
1972
10 ton BOD5/day
95% BOD5
1972
4.5 ton BODj/day
99% BOD5
1975
10 ton BOD5/day
98% BOD5
1975
B-2
-------
REFERENCE LI3T (CONTINUED)
AB Skanebrannerier (Potato Distillery)
Gards Factory
Capacity 2 ton BOD5/day
Reduction 99%
Construction year 1977
3-3
-------
ADDRESS LIST
Swedish Sugar Company
Ortofta sockerbruk
241 00 ESLOV - Sweden
Attention: Plant manager Per Schmidt
Swedish Sugar Company
Hasslarps sockerbruk
260 39 Hasslarp Sweden
Attention: Plant manager Sven Idoff
Swedish Sugar Company
Karpalunds sockerbruk
29100 Kristianstad Sweden
Attention: Plant manager Lennart Rantzow
Swedish Sugar Company
Jordberga sockerbruk
230 20 Klagstorp Sweden
Attention: Plant manager Bengt Lundb.org
AB Skanebrannerier
Box 89
291 01 Kristianstad Sweden
Attention: Managing director Karl - Axel Kornell
8-4
-------
ANAMET PLANTS UNDER CONSTRUCTION IN U.S.
American Crystal Sugar Company
Moorhead, Minnesota
Capacity
Reduction
Contact:
19 ton 30D5/day
95% BOD5
Mr-. Ira Fordycs
3-5
-------
ANAMET PILOT' PLANT' TEST SITES
Universal Foods Corporation
Red Star Yeast Operations
Belle Chasse, LA
Application: Bakers yeast waste
Contact: Mr. James L. Martin
Puerto Rico Distillers Inc.
Arecibo, Puerto Rico
Application: Rum distillery waste
Contact: Mr. Raul Ramirez
Talpe - Star Foods
Kotemark, Belgium
Application: Vegetable canning waste
Contact: Mr. Joseph Talpe
Tirlemont Distillery
Ruisbruek, Belgium
Application: Molasses distillery waste
Fruit Brandy Distillery
Sewen, Switzerland.
B-6
------- |