EPA/600/R-94/169
                                            April 1994


          INNOVATIVE CLEAN TECHNOLOGIES CASE STUDIES

                  SECOND YEAR PROJECT REPORT
              EPA Cooperative Agreement No. CR-817670
                          Project Officer:

                         Kenneth R. Stone
      Waste Minimization, Destruction and Disposal Research Division
                Risk Reduction Engineering Laboratory
                       Cincinnati, OH  45268
This study was conducted in cooperation with the USEPA Office of Small and
                  Disadvantaged Business Utilization
           RISK REDUCTION ENGINEERING LABORATORY
            OFFICE OF RESEARCH AND DEVELOPMENT
            U.S. ENVIRONMENTAL PROTECTION AGENCY
                     CINCINNATI, OH 45268
                                              Printed on Recycled Paper

-------
                                DISCLAIMER
      The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency.  It has been subjected to peer and
administrative review, and it has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.

-------
                                  FOREWORD


      Today's rapidly developing and changing technologies and industrial products
and practices frequently carry with them the increased generation of materials that  if
improperly dealt with, can threaten both public health and the environment. The U.S.
Environmental Protection Agency is charged by Congress with protecting the Nation's
land, air and water resources.  Under a mandate of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance
between human activities and the ability of natural systems to support and nurture life.
These laws direct EPA to perform research to define our environmental problems,
measure the impacts, and search for .solutions.

      The Risk Reduction Engineering Laboratory is responsible for planning,
implementing, and  managing research, development and demonstration programs to
provide and authoritative, defensible engineering basis in support of the policies,
programs, and regulations  of the EPA with respect to drinking water, wastewater,
pesticides, toxic substances,  solid and hazardous wastes, and Superfund-related
activities.   This publication is one of the products of that research and  provides a vital
communication  link between the researcher and the user community.

      This report summarizes projects supported by the RREL to promote the
development of innovative pollution  prevention technologies and techniques by small
businesses.  The reader is encouraged to  contact the individual developers listed in
each summary for more information regarding their process or product.


                            E. Timothy Oppelt, Director
                        Risk Reduction  Engineering Laboratory
                                      in

-------
                                  ABSTRACT


      The Innovative Clean Technologies Case Studies contained herein are the
products of the "Pollution Prevention by and for Small Business" Program (P2SB).  The
P2SB was an outreach program directed to small businesses that had developed
innovative concepts for pollution prevention in their industries.  The P2SB focused on
high-risk concepts without emphasis on media or industry in order to provide an open
program where ground-breaking concepts were given a fair opportunity.

      The P2SB provided awards of up to $25,000 to assist small businesses for
conducting their own demonstrations of pollution prevention techniques and
technologies, and for advancing their products towards a practical stage.

      In its second  year, the P2SB funded projects in a variety of industries across
the nation. This publication provides a history of the  P2SB and lists case histories of
the projects funded  in the second year. A earlier report entitled "Inovative Clean
Technologies Case Studies" (EPA/600/R-93/175) covering the first year of the program
is already  available.
                                      IV

-------
                             TABLE OF CONTENTS


 Phase II and III Vented Two Stage Valves for
 Internal Combustion Engines
       Reggie D. Huff, Aero-Tech, Inc ................................... 1

 Pollution Prevention through Use of a Formaldehyde-Free
 Biological Preservative
       Arthur Schwartz & Barbara Schwartz,  Earth Safe Industries, Inc ......... 18

 Substitution of Biodegradable  Low Toxicity Natural
 Products for the Killing of Fire Ants
       Joe S. Wilkins, Jr. & Joe Wilkins, Sr., Environmental Pesticides Group  ..... 33

 Pollution Prevention in Cadmium Plating
       Mandar Sunthankar, lonEdge Corporation ......................... 55

 Paniculate and Hydrocarbon Emissions Reduction during
 Wood Veneer Drying Operations
       Guy Lauziere, Production Machinery, Inc. &
       Jim Wilson, Oregon State University  ............................. 75

 Conductive Polymer Composites to Replace  Heavy Metals in
 Coatings and Adhesives
       Harry S. Katz & Radha Agarwal, Utility Development Corporation  ........ 91

 Compound Adiabatic Air Conditioning for Transit Buses
       Jamends F. Mattil, Climatran Corporation  ........................  104

 Reducing Heavy Metal Content in Offset Printing Inks
       Roger Telschow, Ecoprint ....................................
Reusing Zinc Plating Chemicals
      Douglas Brothers, Global Plating, Inc .............................  126

In-Ground Plastic Container Production System to Reduce
Nitrate and Phosphate Pollution
      Carl E. Whitcomb, Lacebark, Inc ................................  136

Reuse of Metal Fabrication Wastewaters via a Novel
Ultrasonic Coalescence Process
      Scott R. Taylor, S.R. Taylor & Associates .........................  149

Reduction or Elimination of Cooling Tower Chemicals
      Larry Stenger & Thomas Dobbs, Water Equipment Technologies .......   166

Pre-Charged Vacuum Liquid Extractor/
Containerization Device
      Lowell Goodman, Technical Support Services .......... ............  186

Environmentally Safe Fountain Solutions for the
Printing Industries
      David R. Johns, Summit Resource Management, Inc .................  202

-------
                             ACKNOWLEDGEMENTS


       The Office of Research and Development has supported the P2SB in
 cooperation with the Office of Small and Disadvantaged Business Utilization and is
 grateful for the efforts of Karen V. Brown, EPA's Small Business Ombudsman, who
 and coordinated support for this program.  Through this office the support of the
 following trade associations is acknowledged:

 National Small Business United
 Small Business Legislative Council
 National Automobile Dealers Association
 Independent Lubricant Manufacturers Association
 American Wood Preservers Institute
 Institute of Industrial Launderers
 National Tooling and Machining Association
 Printing Industries of America, Inc.
 International Fabricare Institute
 Synthetic  Organic Chemical Manufacturers  Association
 Chemical  Specialties Manufacturers Association
 National Association of Metal Finishers
 Chemical  Producers and Distributors Association
 National Federation of Independent Businesses
 National Fertilizer Solutions Association
 National Association of Truck Stop Operators
 Society of Plastics Industry, Inc.
 Graphic Arts Technical  Foundation
 American  Galvanizers Association

      The support to small businesses provided by Angel Martin-Dias at the Center
 for Hazardous Materials Research is appreciated. The chapters covering individual
 case histories were written by Ms. Martin-Dias, and CHMR retains copies of individual
 reports.  Support for the commercialization of selected technologies, which was
 beyond the scope of P2SB, was provided by National Environmental Technology
Applications Center (NETAC). The advice and expertise provided by the American
 Institute for Pollution Prevention (AIPP) was  invaluable to the selection process.

      This report was compiled and finalized by Dr. Rada Olbina. Her efforts are
greatly appreciated.
                                      VI

-------
                                INTRODUCTION

      In 1989, the U.S. Environmental Protection Agency established a 2% Set-aside
Program to fund pollution prevention initiatives from across the Agency.  These set-
asides were instituted to encourage the research, development and demonstration of
pollution prevention concepts, techniques and technologies nationwide.

      One such initiative, Pollution Prevention by and for Small Business (P2SB), was
proposed by the Office of Small and Disadvantaged  Business Utilization (OSDBU) with
the support of the Office of Research and Development (ORD).  This initiative was
selected for funding under the pollution prevention 2% Set-aside, with co-funding
provided by ORD.  The P2SB was managed through a cooperative agreement with
the Center for Hazardous Materials Research (CHMR), and some of the P2SB small
businesses received additional support for the commercialization of their technologies
through the National Environmental Technology Applications Center (NETAC).
Nineteen trade associations supported the program though promotion,  advise and
information transfer.

      The P2SB provided awards of up to $25,000 to assist small businesses for
conducting their own demonstrations of pollution prevention techniques and
technologies, and for advancing their concepts to a practical stage. The P2SB was a
three-year program, ending  September 1993, with awards being made in the 1991
and 1992 Fiscal Years of the Federal Government. This publication covers those
projects completed from the 1992 year, and represents the complelation of this
program.

      The P2SB small business applied their own knowledge and expertise in the field
to structure their projects and data collection activities in a manner they determined
would provide the most effective means of furthering the development of their
concepts. The reader should be aware that the data provided in these  summaries are
not the results of third party evaluations.

      The in-house demonstrations were completed  in 14-16 months and reports filed
with CHMR which in turn developed research briefs for Agency review.  These
research briefs have been re-organized into the chapters of this publication. The
technologies are considered promising research and development concepts, and
while several have advanced towards commercialization, others require further
investigation and testing. All are presented to provide the reader the opportunity to
contact the small business demonstrator for potential uses.

      The success of the P2SB  program depends on the involvement of the trade
associations who sponsor presentations by the participating small businesses at
annual conferences  and regional workshops.  This initiative was developed to support
critical pollution  prevention efforts in a variety of facilities and industries that might not
otherwise have the chance or the resources to reduce wastes or to test and
implement their  innovative ideas.  The P2SB was an attempt to support  promising
ventures and encourage further development. It also expanded EPA's knowledge of
pollution prevention  needs in different sectors, supporting government, business and
public cooperation in finding ways to prevent pollution.
                                      VII

-------

-------
                 PHASE II AND III VENTED TWO STAGE VALVES
                     FOR INTERNAL COMBUSTION ENGINES

                                      by

                                 Reggie D. Huff
                                Aero-Tech, Inc.
                               Tigard, OR 97223
ABSTRACT

      Aero-Tech, Inc.'s Vented Two-stage Valving project assessed a concept for
improving the induction process of the internal combustion engine.  The duty of the
induction cycle is to move air and fuel molecules from an inlet venturi to a cylinder
while creating enough turbulence to break up the fuel molecules so they can be
burned efficiently. Unlike standard popet valves, which are of a one piece, one
function design, vented valves are of a two piece, two function design.  This design
facilitates a reduction in mechanical stress and an increased efficiency in air flow
dynamics. The net results are expected to be improved engine performance,
decreased emissions, and enhanced fuel efficiency by creating a more dilute,
homogeneous induction charge, and by allowing better control of induction process
timing.


INTRODUCTION

PROJECT DESCRIPTION
Valve Technology Background

      An internal combustion engine can be described as a sophisticated air pump
that must be able to take in large amounts of air (intake), mix that air with fuel (usually
in liquid form, but sometimes in gaseous form, such as propane) at precise ratios,
continue to move the air/fuel mixture into a combustion chamber, seal itself
completely, compress the mixture, ignite it, thoroughly  extract the energy released
through the resulting expansion of the mixture, and completely purge itself of the
spent gases.  This process must occur in a fraction of  a second. At an extremely
modest operating speed of 2,000 revolutions per minute (RPM), this entire process
must take place within six hundredths of a second or less, and each one of the four
main cycles described has to complete its job efficiently in less than fifteen
thousandths of a second.

      The induction process, which occurs in the first cycle of operation, is the key to
all four cycles working harmoniously and efficiently. The duty of the induction cycle is
to move air and fuel molecules from an inlet venturi to a cylinder within a very short
time; at just the right time; and to fill the cylinder while creating enough turbulence to
break up the fuel molecules to allow efficient combustion.

-------
      The intake valve is always located at the cylinder end of the induction airflow
pattern.  Low pressure is created in the cylinder when the piston, which runs within the
cylinder, begins to descend rapidly down the cylinder on the intake stroke. The
opening of the  intake valve must be timed precisely, opening as expeditiously as
possible, staying open for just the right amount of time, and closing as fast as
mechanically possible. Then, the valve must seal completely and be able to withstand
tremendous heat and pressure.

      Unfortunately, even though the intake valve's job is to allow the induction
system  to flow  as freely as possible within a predetermined time sequence, the head
of the valve itself is the most detrimental obstruction to the air flow pattern.

Basic Flow Dynamics of Vented Valves

      Unlike standard popet valves, which are of a one piece, one function design,
vented valves are of a two piece, two function design.  The main, or outer valve,
controls the flow of the port and is very similar in size and shape to a standard valve
but is designed with  vents in the head portion. This facilitates the need for a  much
smaller  inner valve to control the flow through the vents, as shown in Figures 1 and 2.
This concept allows for a reduction in  mechanical stress.

Figures  1a and 1b.  Standard Intake Valve Flow Dynamics
          FIGURE la
                                               FIGURE  lb
                    FLOW FRONT
                                                STACKING EFFECT CREATES PRESSUR1
                                                IN OPPOSITE DIRECTION OF FLOW

-------
Figure 2.    Vented Intake Valve Flow Dynamics
              LOW PRESSURE CREATED IN CENTER OF
                 VALVE RELIEVES STACKING EFFECT
Figure 3.    Vented Valve Prototype
                               0 "it ft

                                O.IStO
                                                3  *  2
                                                           _L_J,.l
 .	pj

*l     * =

-------
       The vented valve concept follows the basic law of flow dynamics: the further a
 gas or liquid flows away from its given boundaries, the more efficiently it will flow within
 those boundaries. As the inner valve opens, the column of air in the intake port
 begins to collapse inward through the series of vents, past the inner valve, and into
 the combustion chamber. The center 30 to 50 percent of the air column moves first,
 giving laminar flow a more direct passage into the combustion chamber.  This
 alleviates the need for the center of the air column to be radically diverted around the
 head of the valve and into the same space where the outer boundary of the air
 column is flowing. After the center of the air column is moving, the outer valve is
 opened, giving direct flow passage to the outer 50 to 70 percent column of air.  This
 same operation continues in reverse  as the valve mechanism closes.

       The standard valve design  attempts to create air flow that is opposite of normal
 flow dynamics by moving the outer flow first and  preventing the inner laminar flow
 from moving faster and more efficiently through the port.

       Whenever the natural flow pattern of the port is changed,  much of the flow
 dimension  is lost and the performance potential of the entire engine is greatly affected
 (such as with a standard valve).  The vented valve creates a flow that is directly in line
 with natural flow dynamics by creating a multi-layered pathway that taps into higher
 velocity, more efficient areas of the port exit.

 The Dilute Homogeneous Air and  Fuel Combustion Process

       A homogeneous air/fuel mixture is obtained when fuel molecules are completely
 separated and distributed  evenly through a given air mass.  Unfortunately, fuel
 molecules have a tendency to cling together, forming droplets and pools that burn
 and release heat very slowly, as compared to completely atomized fuel.  Since the
 combustion process must occur in such a short time, a slow combustion process
 means that fuel may be unused, unburned, to be expelled into the atmosphere and
 deteriorating air quality.

       Vented valve technology enhances pre-combustion turbulence and creates a
 more homogeneous air/fuel mixture.  When the inner valve mechanism is first
 actuated, an extreme low pressure point is created in the center of the outer valve that
 is surrounded by higher pressure. A  natural phenomenon, the vortex effect, occurs
 and is much like pulling the plug in a  bathtub, only more violent.  This forces the
 center column of air/fuel mixture to move first so it will be continually fed through the
 vents in a high speed  swirling action throughout the operation of the valve
 mechanism.  Creating turbulence directly at and past the intake valve operation is one
 of the most appropriate ways of creating a turbulent, homogeneous air-diluted fuel
 charge, because this event is well-timed with the combustion process.  Vented valves
 tap directly into the high velocity center area of laminar flow, allowing and even
 promoting its high velocity characteristics, rather than acting as a brake or
 perturbation to it. All flow is forced through the center vents at two points in the
 operation of the vented valve, promoting high velocity. Therefore, a new kinetic value
 is seen at the valve, and there is more force, or energy to be diffused into turbulence
 as it enters the combustion chamber instead of creating a high pressure area above
the  combustion chamber.  This is the  only way that turbulence can be increased while
 maintaining or improving flow dimension.

-------
      Some may feel that today's fuel injectors homogenize the fuel molecules
sufficiently such that the aforementioned turbulence is not relevant. Even with throttle
body injectors or central fuel injection, one can observe the speed with which the
finely sprayed fuel molecules pool together and stream down the throttle body walls.
If the injectors are placed directly in front of the intake valve, their  value as
homogenizers is increased; however, the example given shows the value of
turbulence, no matter how or when fuel is introduced into the system.

      Several advantages may be realized by breaking fuel droplets apart by means
of turbulence just before combustion.  Improved flame propagation and velocity allows
the combustion process to take place more quickly. This allows for improved
combustion timing to promote  performance.  More of the combustion process can be
utilized before the exhaust valve opens, which can improve emission quality
dramatically.


Complex Flow Dynamics of Vented Valves

      By reviewing Figures 1 and 2, one can obtain a clearer understanding of how
the vented  valve concept functions in an internal combustion engine.

      If a dormant intake port  instantly begins to fill with  air/fuel mixture from top to
bottom, induced by either thrust pressure or vacuum pressure as the valve is
actuated, then the shape of the flow front that would occur instantly could be
illustrated as shown in Figure 1 a; the center area pushes ahead of the outer  area,
flowing several times faster and more efficiently than the trailing edge of the flow front.
A parabolic lobe-shaped flow front would  appear.  This condition would occur even
with the use of micropolished surfaces. In following the flow front down the port and
closer to the standard intake valve, one can note that the most efficient area  of flow
(i.e., the center of the flow front lobe) will contact the valve base at the center,
therefore having the longest distance to travel along the base of the valve before
entering the combustion chamber. This condition is detrimental to efficient center flow,
so that when the flow front makes contact with the base of the valve (Figure  1b), it
begins to flatten while waiting to enter the combustion  chamber.  When considering
that the flow front is actually a  continuous wave, a stacking effect  can occur on top of
the valve base, as high pressure is created in the center  area of the valve. This
stacking effect pushes the high pressure lobe in the exact opposite direction of the
flow. This provides another opportunity for even atomized fuel molecules to pool
together.

      Although the standard valve design has served well in terms of simplicity,
reliability, and serviceability, it poses the greatest liability to the efficiency of the entire
induction system, even if the system includes a turbo or super charger.  The standard
valve design is simply not allowed enough time to optimize the flow around itself and
continues to deteriorate in efficiency as engine RPM increases.  The vented valve
shows a marked improvement, allowing for more flexibility in cam  profiles with less
overlap for lower emissions and better fuel efficiency, a broader torque and power
band, as well as increased maximum power and torque output levels, before
considering the merit of the forcefully induced swirl effect.

-------
       Aero-Tech identifies "cycle-to-cycle flow kinetics" as a defining factor that
 determines the torque and power curve of a given engine. Cycle-to-cycle flow kinetics
 is based conversely on Newton's Law of Motion.  Thus, the greater the speed of
 motion, the greater the outside force needed to slow that motion.  For example  a
 standard valve engine allows a certain velocity of flow in cycle A. The kinetic energy
 stored in that flow will determine how long the flow will continue to move through the
 port after the valve has closed on cycle A.  This determines how high the pressure or
 charge behind the valve will be when the valve opens for cycle B, and so on  until the
 outside dynamic forces determine the maximum velocity allowed. Components such
 as tunnel rams, increase cycle-to-cycle kinetic energy simply because it takes more
 energy  or outside force, to stop the movement of a tall column of air than a short
 one. Therefore, the air charge is packed tightly behind the valve after every cycle
 propagating velocity. By allowing the very important center column of air to move
 faster and more efficiently through the port, vented valves will increase cycle-to-cvcle
 kinetic energy as well.                                             y      y
 APPLICATION

 Valve Replacement Benefits
       Since vented valves are designed to replace components within the same
 operational space, they would be very cost effective to install, as there would be few
 redesign requirements.  The cost of the vented valve component will be higher than a
 traditional valve because it is slightly more complex, has more moving parts and is
 fabricated from more costly materials such as stainless steel, titanium and '
 molybdenum.   New technology is generally expensive; however cost increases in one
 area can sometimes be offset by the cost savings in another area.

       Enhanced performance, fuel efficiency, and emission quality realized through
 the use of vented valves could allow the auto manufacturer to reduce costs in several
 areas. With improved performance, it may be possible to replace 4  valve-per-cvlinder
 engines with 2 valve engines, thus reducing the number of parts used and simplifying
 production  The use of down-sized  versions of existing engine designs mav also be
 possible  Improved fuel efficiency can save petroleum products. Better emissions
 could reduce the need for expensive emission control  equipment necessary to meet
 s tnngent emission standards. For example, catalytic converters, containing costly
 platinum as a catalyst, are used in exhaust systems to burn-off unburned

                                ^ impr°vement  in Pre-cata|yst exh*-< oo* result
Cross Segment Uses

TK  4. Cfveral1 marke! P°tential for vented valve technology is quite large and diverse
The twelve mam market segments, in no significant order, are:              u'vwi&e.

      Automotive: original equipment manufacturers (OEM)
      Truck: light, heavy duty, semi  (OEM)
      Industrial: machine power, generators, etc. (OEM)
      Engine rebuilders: auto, truck, etc., aftermarket manufacturers (AM)
      Engine modifiers: high performance, racing (AM)

-------
      Defense
      Government: federal, state, local fleet, etc.
      Aviation: (OEM), (AM)
      Marine: (OEM), (AM)
      Recreational vehicles: motorcycles, etc. (OEM), (AM)
      Agriculture: (OEM), (AM)
      Locomotive

      The automotive and light and heavy duty truck segments have the most to gain
from successful development of vented valve technology. However, the cost of
implementing new technology is relative to the benefit gained.  Clear 10 percent
improvements in fuel efficiency, emissions, and performance may be worth a cost
increase of 20 percent.  However, a 1 percent improvement might be considered to
expensive at a 5 percent increase in costs.
PROCEDURE

DEMONSTRATION SET-UP

      The engine selected for the project was a 1989 Mitsubishi 2.0 liter, 4 cylinder,
single overhead cam with computer controlled, multi-point sequential port fuel
injection.

      The engine was removed intact from the automobile, and the accompanying
wires and electronics were mounted on a panel to simplify operation.  A special
bellhousing with an adapter  plate and a special clutch disc were made in order to
adapt the motor to the SF 901 Dynamometer.

      The motor was completely dismantled, inspected, and found to be in excellent
condition. The pistons were removed and .075 inch deep relief pockets were
machined in order to ensure that the inner valves of the vented valve units would have
sufficient room to  function without making contact with the pistons.

      All relevant data were gathered while the engine was disassembled.  (For
example, the compression ratio of 7.8:1 was obtained using a cc burette.) The engine
was reassembled  and run, free load, on the engine stand. When all systems were
determined to be  functioning properly, the test engine was ready for the standard
valve baseline testing.

      A number of problems were encountered during the initial machining of the
vented valves.  The incorrect Rockwell hardness rating was provided for the stainless
steel blanks used for the outer valves, which  resulted in numerous delays. The inner
titanium valves were machined easily.

      Once the valve pieces were completed, the inner valves were lapped-in  for
proper sealing, and the springs and pins were adjusted and installed.

-------
 Engine Testing

       Engine tests were conducted at Western Washington University, Bellingham
 Washington, at Dr. Michael Seal's Vehicle Research Institute.  Aero-Tech rented the
 dynamometer laboratory and mounted the engine in stock form of the SF 901
 Dynamometer.

       An exhaust probe from an exhaust gas analyzer was set up to track
 hydrocarbons and carbon monoxide emissions throughout the tests. Unfortunately
 the NOx analyzer was malfunctioning at the time of the test and could not be repaired
 in time.  Aero-Tech believes that these  exhaust gases would not be problematic
 because exhaust gas temperatures dropped during the tests.

       The exhaust emissions analyzer  required a separate process to record the data
 manually as the dynamometer runs were conducted.  This manual recording proved to
 be inefficient, and therefore the emissions data are not a thorough and precise as the
 dynamometer data.  However, definite trends were revealed, and other factors must
 be analyzed in the context of real-life situations to determine the potential gain or loss.

       Tests were conducted with a dynamometer, an instrument designed to analyze
 an engine's ability to resist precisely measured loads and calculate horsepower and
 torque. Today's dynamometers are computerized and can extract and analyze a
 great deal of data on such parameters  as fuel efficiency, volumetric efficiency and
 thermodynamics.

       The Mitsubishi 2.0 liter, 4 cylinder ran well during  the tests.  Dynamometer runs
 over 6,000 RPMs were made without any problems.  Once the data had stabilized
 over several dynamometer runs, Aero-Tech collected all  relevant data from the stock
 engine.

       The standard valves were then removed and the vented valves installed
 Several special reamers  enlarged the valve guides from  8 millimeters to 9 millimeters
 to accommodate the stem of the vented valves.  The cc burette was  again used to
 measure the volume of the combustion chamber, and the engine was reassembled.

       The vented valve engine first  was run slowly below 2,500 RPMs and was
 gradually increased to beyond 6,000 RPMs.


 EVALUATION PARAMETERS

      All data gathered on both engine configurations were corrected to Society of
Automotive Engineers (SAE) standards, which mathematically corrects the ambient
atmospheric conditions observed at the time of the test to a standard of 100
kilopascals absolute (29.38 inches mercury) inlet air pressure and 25 degrees
centigrade (77 degrees fahrenheit) inlet air temperature.  Dr. Michael  Seal, Director of
the Vehicle Research Institute, witnessed and certified the data, which include
comparisons of horsepower, torque, brake special fuel consumption (BSFC) brake
specific air consumption  (BSAC), and air to fuel ratios.
                                      8

-------
ASSESSMENT SUMMARY

      An analysis of various factors revealed that Aero-Tech had not achieved an
optimum assessment of the technology's potential.  After solving some minor timing
problems, the modified vented valve engine started and ran smoothly with no
noticeable increase in noise or vibration.  An objective analysis also revealed an
improved throttle response.


DEMONSTRATION COST

      The total cost of the demonstration was $48,042.72, with $25,000 being covered
by EPA's Pollution Prevention By and For Small Business program, and the balance of
$23,042.72 shared by Aero-Tech.


RESULTS AND DISCUSSION

TEST RESULTS

      Project results indicate that vented valves do work and have a great deal of
potential.

      All critical data were tabulated, sorted, and organized by the dynamometer
computer at any RPM point chosen by the dynamometer operator.  This results in raw
data from various RPM points.  In order to plot evenly spaced points (e.g., 500, 1,000
1,500 RPMs), the data need to be interpolated using  a mathematical formula:

            Y2 = Y1 + (((Y3 - Y1) -f (X3 - X1))(X2 -  X1))

where:
      X2 equals the desired evenly spaced data or RPM point
      Y2 equals the unknown value for that point
      X3, X1, Y3, and Y1 represent raw data points from the dynamometer
      This yields credible data at evenly spaced data points that allows for graphing
 and making easy comparisons to other data within the project, and to data from other
 projects.

      Torque and  horsepower data for the stock engine and the vented valve engine
 are shown in Table 1 and Figure 4.

-------
TABLE 1.    Normalized Torque and Horsepower Date for Modified (Vented Valve)
           and Stock Engines

RPM
1,700
2,000
2,500
3,000
3,500
4,000
4,500
5,000
5,500
6,000
Normalized Torque Data
(R/Lbs)
(SAE Net)
Vented
101.90
100.90
101.85
106.50
102.42
103.98
103.15
101.47
93.83
80.95
Stock
67.92
66.92
67.00
71.36
73.66
77.74
81.00
79.96
74.92
67.37
Delta %
50.01619
50.76715
52.01624
49.237
39.03991
33.75204
27.35045
26.89902
25.24843
20.16507
Normalized Horsepower Data
(SAE Net)
Vented
33.00
37.50
42.50
54.19
66.49
78.94
81.55
89.69
91.97
88.52
Stock
22.01
25.48
32.84
40.92
49.08
59.42
69.39
76.10
78.05
76.92
Delta %
49.89231
47.16608
29.41703
32.41595
35.47362
32.84486
17.52843
17.85621
17.84094
15.0893
                                   10

-------
Figure 4.    Torque and Horsepower Data
                         TORQUE/HORSEPOWER DATA
               Torqiw (Ft/lbs) SAE Nat
• MOO
0 STOCK
                                            Hor»«pow*r (SAE tot)
• MOO
0 STOCK
    60

    50

    40

    30

    20

     10

      0*
                 DELTA (%incr«a»«)
i:  i, -
          SSStttwsM
I  !  I  IillLJ_U
                                                       DELTA (% IncrMM)
SO
45
40
36
30
25
20
16
10
      i   a  a
1  *  i   I  8
i  i  I   S  i
  i  •  X  1   •  I  •_•_••
       A comparison of air/fuel consumption for the stock engine and the vented valve
 engine is shown in Table 2 and Figure 5.
                                      11

-------
 TABLE 2.
Normalized Air/Fuel Consumption Data for the Modified (Vented Valve)
and Stock Engines

RPM
2,000
2,500
3,000
3,500
4,000
4,500
5,000
5,500
6,000
Normalized BSFC Data
(Lbs/Hp.Hr)
Vented
0.61
0.53
0.56
0.62
0.65
0.67
0.69
0.72
0.80
Stock
0.64
0.67
0.67
0.63
0.62
0.70
0.74
0.76
0.86
Delta %
-5.98593
-20.4018
-15.8277
-1.14517
5.668204
-4.27512
-7.44707
-5.80395
-6.43653
Normalized BSAC Data
(Lbs/Hp.Hr}
Vented
14.09
14.10
13.45
12.57
10.34
10.17
9.54
9.59
9.97
Stock
8.16
7.73
8.08
7.27
6.91
6.87
7.19
7.86
8.50
Delta %
-42.1091
-45.1744
-39.9009
-42.1411
-33.1594
-32.4564
-24.6909
-18.041
-14.67
BSAC: Brake Specific Air Consumption
      A comparison of hydrocarbon and carbon monoxide emissions data for the
stock engine and the vented valve engine is shown in Table 3 and Figure 6.

TABLE 3.    Normalized Hydrocarbon and Carbon Monoxide Emissions Data for
            Modified (Vented Valve) and Stock Engines

RPM
2,000
2,500
3,000
3,500
4,000
4,500
5,000
5,500
6,000
Normalized Hydrocarbon Data
(ppm)
Vented
265.48
186.11
193.22
190.87
177.44
197.98
200.00
200.00
200.00
-
Stock
202.27
216.48
229.59
241.06
256.96
288.61
316.67


Delta %
31.24666
-14.0274
-15.8401
-20.8189
-30.9456
-31.4025
-36.8421

=====
Normalized Carbon Monoxide
Data (%)
Vented
3.00
3.00
3.00
3.20
3.45
3.96
4.67

=====
Stock
4.45
4.59
4.72
4.85
5.00
5.00
5.00
4.82
4.61
-
Delta %
-32.5843
-34.627
-36.4407
-33.94
-30.9118
-20.809
-6.56371

===—=—
                                    12

-------
Figure 5.    Air/Fuel Consumption
                                  AIR/FUEL CONSUMPTION
       Brake Specific Fuel Consumption   0bWhp.hr)
• MOO
0 STOCK
Brake Specific Air Consumption (Iba/hpJhr)
                                                    1600
                                                    14.00
                                                     000
               BSFC Delta (% decrease)
• MOD
0 STOCK
                                                                  BSAC Delta (%decreaeey
                                           13

-------
Figure 6.     Emissions Data
                                  EMISSIONS DATA
          Hydrocarbon Emission* (PPM)
• MOO
0 STOCK
Carbon Monoxlds Emission* (%)
• MOO
0 STOCK
              DELTA (* Chang* W-)
                                                        DELTA (X Chang* */->
                                       14

-------
      Reviewing this data illustrates the profound effect vented valves had on the test
engine.  Special attention should be given to the torque data, which indicates that an
element of variable lift and timing was present, as a very flat torque curve was created
(the flattest observed by Dr. Seal and Mr. Huff).  The implications of this observation
suggest the possibility of using lighter transmissions with fewer gears, giving vehicle
designers another tool to increase efficiency.

      The brake specific fuel consumption and brake specific air consumption are
measured in pounds per horsepower hour, and are a direct gauge of the efficiency of
the engine, indicating how much power the engine is producing relative to the amount
of air and fuel consumed.

      The emissions data are pre-catalyst; they represent the actual emissions
measured without a catalytic converter  or air pump applied to the exhaust system.
The emissions were measured under fully loaded dynamometer testing.

FACTORS IMPACTING PERFORMANCE

      One concern of using a late model engine with computer controlled ignition and
fuel delivery systems was that if this technology exceeded the parameters
programmed into the data computer, the performance results could be skewed. This
apparently  happened with the test engine. The dynamometer's computer indicated
that the standard valve engine ran with better air/fuel ratios than the vented valve
engine. This was caused by the tremendous increase in the  vented valve's flow
dimension and turbulence that "confused" the computer and/or sensors into thinking
that the engine was running too lean (i.e., there was too much air to the fuel, or the air
to fuel ratio was too high).  The computer then increased the fuel flow, which actually
resulted in  the engine  running far to  rich  (i.e., the ratio of air to fuel was too low).

Air/Fuel Ratio Comparison

      The  air/fuel (A/F) ratio comparison for the stock engine and the vented valve
engine is shown in Table 4.

TABLE 4.   Air/Fuel Ration Comparison for the Stock and Vented Valve Engines
RPMs
1,500
2,000
2,500
3,000
3,500
4,000
5,000
5,500
Stock A/F
16.7:1
16.0:1
15.4:1
14.7:1
13.4:1
13.1:1
11.9:1
12.2:1
Vented A/F
12.3:1
12.5:1
12.8:1
13.7:1
11.0:1
10.4:1
10.5:1
10.5:1
Delta % Increase
-.357
-.28
-.20
-.072
-.21
-.25
-.13
-.16
Mean average % decrease in air/fuel = .20
                                       15

-------
       The modified vented valve engine ran at a distinct disadvantage to the standard
 valve configuration.  The vented valve engine, with its increased velocity and
 turbulence, could have run leaner than the standard valve engine.  Yet it outperformed
 its counterpart while running too rich. All data, especially concerning fuel efficiency
 and emissions, would have improved significantly had Aero-Tech been able to adjust
 the air/fuel ratio to optimum levels.
 FAILURE ANALYSIS
       Upon completion of all tests, the top of the engine was dismantled and the
 vented valves removed to inspect their condition following the testing. The vented
 valves were in perfect condition, with no abnormal wear, cracks, or blemishes
 However, one of the inner valves was stuck fully closed, apparently caused by a piece
 of dislodged machining debris getting caught between the inner valve stem and guide
 Aciro-Tech cannot determine whether the valve was stuck throughout the test.


 COST/BENEFIT ANALYSIS

       Based on the data gathered, Aero-Tech believes that a 20 percent fuel savings
 could be realized with the vented valve engine over existing designs  Using 1989 U S
 new car sales, the approximate fuel savings can be calculated using a 1  percent 5 '
 percent, and 10 percent vented valve implementation. Reductions in  liquid fuel  '
 consumption (cost savings to the consumer) are calculated to be:

       1  percent = 11,840,922 gallons saved
       5 percent = 59,204,610 gallons saved
       10 percent =  118,409,220 gallons saved

       For the manufacturer, vented valves offer the possibility of less mechanical
 complexity in order to obtain desired fuel efficiency and emissions standards  as well
 as the potential of requiring less exhaust catalyst and perhaps smaller, less expensive
 transmissions.                                                          M


 CONCLUSIONS

 POLLUTION PREVENTION ASSESSMENT

      Pollution prevention is largely associated with fuel savings. If two automobiles
 of equal weight and size burn fuel at totally different rates, it is likely that the auto
 burning the most fuel also emitted the most pollution.

      If vented valves lead to the use of  smaller engines and  transmissions this
simplified manufacture could reduce pollution produced in the manufacturing process.

      Currently, vented valve development is in the third stage of development and is
nearmg production of an aftermarket product.  This will lead to extensive failure
analysis and durability testing that could lead to an in-market production vehicle
product.
                                      16

-------
      Industry resistance to new product design from outside sources may prove to
be a barrier to the acceptance of vented valves.  Further testing and refinement to
generate replicable data, as well as its success in aftermarket sales, can promote the
adoption of this technology.
                                       17

-------
                 POLLUTION PREVENTION THROUGH USE OF A
               FORMALDEHYDE-FREE BIOLOGICAL PRESERVATIVE

                                      by

                    Dr. Arthur Schwartz and Barbara Schwartz
                            Earth Safe Industries, Inc.
                             Belle Mead, NJ 08502
ABSTRACT
      Formaldehyde, a major component of most biological preservatives, is a toxic
air pollutant.  67 billion pounds of this substance are produced in the U.S. annually.
Earth Safe Industries' product,  NoToX, is a non-toxic, biological preservative intended
to replace formaldehyde in a variety of applications. This project demonstrated the
effectiveness of NoToX in eliminating air pollution (associated with the venting of
formaldehyde) and acting as an effective, long-term preservative and fixative.


INTRODUCTION

PROJECT DESCRIPTION

      Earth Safe Industries' objective for this project was to demonstrate that NoToX
Biological Preservative and NoToX Histological Fixative could perform preservation
and fixation tasks equivalently to formaldehyde in the biological and medical sciences.
These tasks include anatomical embalming (needed for the study of gross anatomy),
fixation and preservation of whole organs (used in research and autopsy
investigation), and fixation and preservation of large tissue blocks and thin tissue
sections  (used to determine the absence or presence of disease states from surgical
patients). Since NoToX Biological Preservative and NoToX Histological Fixative are
made from non-toxic and non-carcinogenic constituents, pollution can be eliminated
from the  source. Approximately 5 million gallons of toxic waste is generated from
formaldehyde use in the  biological sciences annually; simply replacing NoToX where
formaldehyde has traditionally been used eliminates this pollution.  Earth Safe
Industries' extensive testing of NoToX to meet the criteria of these applications found
that NoToX is capable of replacing formaldehyde in all these protocols, and that
researchers and health professionals can rely on NoToX to perform protocols equal to
or better  than formaldehyde.  The data obtained from this study were instrumental in
convincing formaldehyde users that performance would not be .sacrificed for safety.
NoToX Biological Preservative and NoToX Histological Fixative have made their way
into the marketplace throughout the U.S., Canada, Europe, Mexico, and South
America.

Product Outline

      NoToX Biological Preservative is an environmentally safe anatomical embalming
fluid developed specifically to meet the stringent requirements of anatomical
preservation,  both human and veterinary.  Excellent long term  anatomical preservation
is achieved, and importantly, chemicals that are carcinogenic, toxic, and corrosive


                                      18

-------
(such as formaldehyde, phenol, and glutaraldehyde) have been completely eliminated.
NoToX is formulated as an aqueous alcoholic solution consisting of protein stabilizers
and modifiers, humectants, and potent antiseptic and antifungal agents.  These
constituents are regarded by regulatory agencies as environmentally and
occupational^ safe. With the absence of toxic and carcinogenic substances in the
biological preservative, laboratory personnel can work safely with NoToX for unlimited
periods of time.  There is no irritation to the eyes and respiratory tract when personnel
work with NoToX, nor is there the offensive, heavy, acrid odor commonly associated
with formaldehyde and phenol.  On the contrary, NoToX has a mild aroma and is a
pleasant substance with which to work.

      NoToX Histological Fixative has the long term preservation features  of the
biological preservative, but has been modified to meet the needs of histological tissue
processing. Specimens prepared for histological evaluation are smaller  and  require
less humectants, bactericides, and fungicides. The product does not sacrifice safety
and performance, and has been optimized to meet this  application traditionally
delegated to a 10 percent neutral buffered formalin.

      NoToX Histological Fixative features a  unique chemical system that stabilizes a
bis carbonyl compound  and activates it towards a crosslinking reaction with  protein.
The NoToX crosslinker is selective and does  not attack  antigenic sites or DMA.
Consequently superior results are obtained with  immunphistochemical stains and in
situ hybridization techniques.  Although NoToX Histological Fixative can  replace
formaldehyde completely, NoToX Histological Fixative is interchangeable with formalin-
fixed specimens.  Formaldehyde-fixed specimens arriving at a laboratory can be
placed in a tissue processor containing NoToX and over-crosslinking will not occur.
Thus, laboratories obtaining specimens from  sources outside their facility need not
disqualify their laboratory from using NoToX.  Excellent results can be achieved when
a specimen has initially been fixed in formaldehyde and  then processed  in NoToX
Histological Fixative.

Unique  Product Features

      The  elements of a common formula that produce superior tissue fixation unite
the anatomical and histological applications.  Key to the efficacy of NoToX Biological
Preservative and NoToX Histological Fixative  is a unique bis carbonyl compound that
has the ability to crosslink protein in a manner similar to formaldehyde.  The  bis
carbonyl compound is dissolved in an aqueous, alcoholic solvent and is fortified with
high potency, non-toxic, antimicrobial agents that prevent the proliferation of bacterial,
fungal, and viral pathogens.  Alcohol not only functions  as a supplementary protein
denaturant, but also has the solvency power to dissolve certain lipophilic, germicidal
compounds at much higher levels than could be attained in a purely aqueous
medium. A polyhydridic alcohol, such as glycerin  or propylene glycol, is included in
the formulas as a humectant. The humectant is  important in the anatomical
application, as it retards the dehydration of unsubmerged or exposed tissue.  In the
histological application, unsubmerged tissue  is never a problem. Since  the
polyhydridic alcohol has little, if any  overall impact on the degree of protein
denaturation, it can be omitted from the histological application with almost no loss of
function. The polyhydridic alcohol has not been omitted from the histological formula,
but has been reduced in concentration.
                                       19

-------
 APPLICATION

 Products Replaced

       Four products emerged as a result of Earth Safe's studies.  These are NoToX
 Biological Preservative and NoToX Histological Fixative, both of which replace
 formaldehyde and other toxic components used for preservation and fixation in the
 medical and biological sciences. Accessory products that are used in conjunction
 with anatomical preservation were also developed to replace current toxic
 counterparts. As the investigation was conducted, embalmers stated that in addition
 to the use of formaldehyde, either phenol or paraformaldehyde crystals were used to
 retard fungus formation on cadavers during storage.  Some institutions had
 refrigeration facilities and wrapped cadavers in plastic bags that were stored in cold
 rooms.  Paraformaldehyde crystals were sprinkled into the bags to inhibit fungus
 formation on the specimens.  To replace these crystals and eliminate this source of
 pollution, an  environmentally safe, non-toxic, non-carcinogenic, antifungal compound
 was developed with the trade name, VaporSafe.  TankGuard was developed for those
 institutions using tanks for cadaver storage.  These tanks, ranging from 30 gallons to
 5,000 gallons, hold cadavers until needed.  Phenol is added to the tanks to retard the
 formation of fungus.  The environmentally safe TankGuard replaces the toxic phenol in
 this application. The development of these four products eliminated toxins off-qassed
 by formaldehyde, phenol, and glutaraldehyde.

       NoToX Biological Preservative also proved to be an excellent replacement for
 formaldehyde in preserving dissection specimens (e.g., frogs, fish, cats, and rats)
 used in many biology classrooms.  A major supplier of preserved animals successfully
 tested NoToX Biological Preservative as a replacement for its
 formaldehyde/glutaraldehyde fluid.  Because of a downturn in the preserved animal
 market, the company opted not to adopt NoToX as its preservation fluid.  Several
 schools of veterinary medicine (including Tufts University and the University of
 Montreal Schools of Veterinary Medicine) have adopted NoToX Biological Preservative
 to embalm cows, horses, goats, dogs, and cats for anatomical study.

       NoToX Histological Fixative replaces formaldehyde in fixing and preserving
 biopsy specimens excised from  patients or whole organs derived from autopsy
 examination, and formaldehyde as used in tissue  processors in hospital and private
 medical laboratories.   In these applications, approximately five million gallons of toxic
 formaldehyde waste can be eliminated annually.

 Wastes Prevented

      Formaldehyde, phenol, glutaraldehyde, and toxic substances off-gassed by
 paraformaldehyde crystals are eliminated at the source with the use of NoToX
 Biological Preservative, NoToX Histological Fixative, VaporSafe, and TankGuard.

      NoToX Biological Preservative has the capability of eliminating 100,000 gallons
of formaldehyde and  phenol waste generated annually by medical, dental, and
veterinary schools from anatomical embalming. Another source of formaldehyde
pollution is dialysis machines.  Approximately 100,000 gallons of formaldehyde are
used to disinfect components from dialysis machines annually.  Since NoToX
                                      20

-------
Biological Preservative has potent disinfecting capabilities, it can replace formaldehyde
in this application and eliminate the waste formaldehyde generated.

      Approximately 5 million gallons of formaldehyde waste is generated annually in
the U.S. from the use of tissue processors used to prepare tissue samples for
microscopic evaluation.  Hospitals and private medical laboratories can save as much
as $140 million annually  on waste disposal costs simply by switching to NoToX
Histological Fixative and eliminating the generation of the formaldehyde waste.

Cross Segment Uses

      As a formaldehyde substitute, potential cross industry uses for NoToX include:

•     Veterinary, medical, and dental schools that embalm human and animal
      cadavers for anatomical study in the training of health professionals.  NoToX
      Biological Preservative is presently being used for this application.

•     Medical examiners' offices  perform autopsies and  preserve whole organs in
      large quantities of formaldehyde. NoToX Biological Preservative can preserve
      whole organs  by passive immersion better than formaldehyde, as NoToX's
      selective crosslinking mechanism permits infiltration into the core of the  organ.
      Formaldehyde tends to fix specimens on the outer parts of the organ since  its
      crosslinking mechanism is random, which means that the core of the organ  --
      particularly whole brains - is not always fixed.

•     Preservation of animals and specimens for dissection may be achieved with
      NoToX Biological  Preservative. Testing was done with NASCO, a major
      worldwide supplier of preserved specimens  to schools, universities, and
      research institutions that validated the utility  of NoToX Biological Preservative in
      this application.

•     NoToX Biological  Preservative may be applied as an embalming fluid for the
      commercial funeral industry.  NoToX's long term preservation capabilities
      nave already been proven in medical and veterinary schools and may meet  the
      needs of the commercial funeral industry.

•     In the dialysis industry, NoToX  Biological Preservative may be tested to
      replace formaldehyde to disinfect dialysis components.

•     Hospitals, private medical laboratories, and research institutions may be
      able to eliminate formaldehyde in selected applications for tissue processing.
      NoToX Histological Fixative performs on a par with formaldehyde that is
      diagnostically equal as shown in  the histological  portion of the project at Mount
      Sinai Medical Center.
                                      21

-------
PROCEDURE

DEMONSTRATION

      The NoToX project was divided into three categories: Anatomical,
Histological, and Microbiological.

Anatomical Demonstration

      The anatomical phase of the  project was conducted under the direction of Mr.
Roger Faison,  Principal Investigator and Supervisor of the Gross Anatomy Department
at the University of Medicine and Dentistry of New Jersey in Newark. Cadavers were
embalmed with NoToX Biological Preservative and allowed to incubate under
refrigerated storage conditions for varying periods of time:  1, 3, 6, 9, and 12 months.
The objectives were to assess preservation and to determine if minimum and
maximum time requirements existed for achieving good preservation. At the end of
the incubation  period, effectiveness  of preservation was demonstrated by having
students dissect the cadavers in their usual manner.  Examination of internal and
external regions of the cadavers would indicate effectiveness of the NoToX.  In some
cases, cadavers were stored at room temperature to determine whether NoToX would
be effective in institutions without refrigeration facilities.

Histoloqical Demonstration

      Dr. Pam  Unger of Mount Sinai Medical Center's Department of Pathology  in
New York City served as Principal Investigator for the histology section.  Fifty one
surgical specimens were fixed in neutral buffered formaldehyde and NoToX
Histological Fixative and  processed  using standard histological protocols that were
established for their laboratory.  For all surgical cases, protocols for formaldehyde-
fixed tissue and NoToX fixed tissue  were identical.  After slides had been prepared,
the tissue samples were  evaluated microscopically and compared.

Microbiological  Demonstration

      As a subsection of the Anatomical investigation, blood was drawn from
cadavers before and after embalming. The blood was cultured  relative to a
representative panel of bacterial  and fungal organisms.  Colonies were counted in the
pre- and post-embalmed sarpples to determine the reduction in inhibition of the
organisms by NoToX. In another part of this study, which was conducted as part of
the optimization procedure,  microbial inhibition  of NoToX was compared to
formaldehyde and phenol by means of the Minimum Inhibitory Concentration and
Kirby Bauer disc inhibition methods.

EVALUATION PARAMETERS

Environmental

      While formaldehyde, phenol,  or other toxic substances are not used in the
NoToX formulations, Earth Safe elected to determine that no off-gassing of
formaldehyde occurred through decomposition, or any other means. Unlike
formaldehyde, there is no possible degradative mechanism by which phenol could


                                      22

-------
evolve.  Levels were assessed merely as a point of validation, since phenol is always
used together with formaldehyde in conventional anatomical embalming compositions.
In the Anatomical Demonstration, where large amounts of the fluid were used for
embalming cadavers, airborne concentrations of formaldehyde, phenol, and ethanol
were measured with conventional air sampling equipment.  Ethanol was sampled
because NoTpX is formulated with ethanol, and there is a remote chance that under
certain conditions, levels may become elevated.  In the Histolpgical Demonstration,
where relatively small amounts of the NoToX Histological Fixative are handjed by
technicians, air sampling is not necessarily a preferred  method for monitoring levels of
formaldehyde. The accepted method for monitoring personnel who are exposed to
formaldehyde, in this case, is by requiring them to wear passive dosimeter badges.
This was done for the personnel who were working with NoToX in the histology
laboratory even though the likelihood of observing an elevated value was remote.
Since ethanol levels in the histology laboratory are never considered a health threat,
they are not measured.  Therefore, they were not measured for NoToX either.  Since
phenol is not used in the histology laboratory, levels were not determined. Any level
of phenol, regardless of how low, could be detected without instrumentation, although
quantification would not be possible.  None of the technicians or other handlers of
NoToX reported the detection of phenol. Preclusion of the  determination  of phenol
concentrations in this case, is justifiable on the grounds that the action  level for phenol
is 2.5 ppm, but the odor threshold is 0.04 ppm.  Phenol can be detected  by  smell well
before it rises to a danger level.

Anatomical

      Preservation efficacy of the embalmed cadavers  was judged by the following
criteria:  internal and external  color of the cadavers; general  appearance; firmness of
internal  and external regions, including the  abdominal area  and internal organs; ease
of separating and identifying tissue; ability to identify fascial planes; and ability to
maintain the integrity of muscle tissue.

Histological

      The most important criteria for determining the effectiveness of NoToX
Histological Fixative was that it provide diagnostic results that were equivalent to
formaldehyde. Results that were too divergent, even if  superior, might not be
acceptable.  Parameters used for judging NoToX Histological Fixative's utility were
histologic criteria -- ease of processing and cutting specimens -- and microscopic
criteria -- stain intensity for both conventional and immunohistochemical stains, cellular
architecture, nuclear detail, and contrast.

Microbiological

      The reduction in organism colonies after embalming was the key parameter in
the  microbiological study.  With respect to the two other microbiological studies
previously mentioned, the objective was to  compare the potency of NoToX Biological
Preservative to formaldehyde and phenol-based  preservatives.
                                      23

-------
ASSESSMENT SUMMARY

Environmental

      As part of the Anatomical Demonstration, assessment of airborne levels of
formaldehyde, phenol, and ethanol were determined with conventional air sampling
equipment from "Alltech."  Background levels of formaldehyde had to be determined
because cadavers that had been previously embalmed with formaldehyde were being
stored in the same general area of the anatomy laboratory as were the NoToX-
embalmed cadavers.  In general, the NoToX cadavers were kept as far from the
formaldehyde/phenol cadavers as possible, but it was impossible to prevent some
commingling of the vapors.  Sample air measurements were made by keeping the
probe about one foot from the cadaver at a point approximately 1.5 feet above the
platform of the dissection table. Background measurements were made in the
laboratories on a day when no embalming or dissection  activity of
formaldehyde/phenol cadavers was being conducted. For background
measurements, sampling probes were positioned with respect to cadavers as they
would be when actual embalming or dissection activity was taking place.  Since the
entire laboratory area was permeated with the odor of phenol, background
measurements had to be made with the same consideration as was explained above
for formaldehyde.

      Ethanol is not used in the UMDNJ laboratories. Their normal procedures call
for large quantities  of isopropyl alcohol.  Ethanol is non-toxic with a TLV of 1,000 ppm,
but isopropyl alcohol  is a toxic substance with at TLV of  400 ppm. Earth Safe's
concern was with the specificity of the sampling equipment. Earth Safe felt, that
despite the manufacturer's claims, the chemical similarity of ethanol to isopropyl
alcohol would tend to give false positive values for  airborne ethanol.  Background and
sample levels of ethanol turned out to be so low and so  far from the TLV value, that
concerns over specificity were unfounded.

      Measurements of the above substances were made in the laboratory used for
embalming as well as that used for dissection (student gross anatomy lab).  Levels of
formaldehyde were always at or below background levels (0.6 ppm) in all critical areas
where NoToX was utilized.  The permissible exposure limit  (PEL) for formaldehyde is
0.75 ppm. Airborne concentrations of phenol were below background levels (less
than 1 ppm)  in all critical laboratory areas where NoToX  was utilized. The PEL for
phenol is 5 ppm. Background and sample values for ethanol were about 0.3 and  13
ppm respectively, a small fraction of the 1,000 ppm TLV  value.  Typical data that were
collected for  air sampling in different areas are shown in  Table 1 in the Tabulation of
Data section.

      Formaldehyde dosimeter badges from "Alltech" were worn by technicians who
were working with formaldehyde during the Histological Demonstration. Values of
formaldehyde were always less than the background value, which was 0.2 ppm.

Anatomical

      No difference in preservation ability was noted for  specimens stored at the 1, 3,
6, 9, or 12 month intervals.  Fixation occurred rapidly so  that good preservation was
observed before 1 month, and that long term preservation was of sufficient quality so


                                     24

-------
that a well produced specimen was still retained after 12 months of storage under
refrigeration conditions.  Likewise, cadavers stored at room temperature were of
preservation quality equal to those stored in the cold.  The overall appearance of the
preserved cadavers was quite good.  External colors had a pale yellow, natural
appearance in contrast to the grey-brown color of cadavers embalmed with
formaldehyde/phenol solutions.  Extremities were well-preserved, and mold and fungal
growth on cadavers was not observed. Textures were firm but not dehydrated or
mummified as would be the case if over-crosslinking had occurred.  In particular, the
abdominal region felt firm on all specimens. Upon dissection, the integrity of muscle
tissue appeared well-maintained, tissues were easy to separate and identify, and
fascial planes were readily identifiable. Internal  organs, such as the heart and
intestines, were firm.

Histoloaical

      The significant conclusion drawn from this study was that NoToX Histological
Fixative is diagnpstically equal to formalin.  Tissue fixed in NoToX Histological Fixative
showed no significant differences either in the processing of tissue  or in the  cellular
morphology as judged by the criteria above.  Optimal results were  obtained  without
vacuum in the fixation chamber.  NoToX Histological Fixative showed weaker staining
of cytoplasmic granules, but otherwise, cytoplasmic staining was comparable.  Eosin
and immunphistochemical stains showed greater stain intensity, but diagnostically this
is not perceived as being a problem,  since in the case of eosin, the greater stain
intensity makes for greater contrast.   NoToX Histological Fixative, in contrast to
formaldehyde, does not destroy antigenic sites, which allows for greater stain intensity
in immunohistochemical stains. Since there is a greater amount of  antigen available
for reaction with antibody, the immunohistochemical stain becomes too intense and
high background  stain results. Rather than changing the fixative formulation, a better
approach would be to instruct the user to titer the antibody concentration for optimal
contrast. By  using 25 percent less of the amount of antibody normally used for
formaldehyde-fixed tissues, superior immunohistochemical stains can be achieved.
This practice  saves money for the user by reducing the amount of antibody  needed
for immunohistochemical staining.

Microbiological

      Fluid withdrawn from a cadaver that had been embalmed for one month
showed zero population colonies for a panel of bacterial and fungal organisms
including E. coli, P. aeruginosa, S. aureus, A. niger, and C. albicans.  A microbiological
study conducted  by Earth Safe Industries was designed to generate comparative
minimum inhibitory concentration (MIC) values for NoToX Biological Preservative,
formaldehyde, and phenol.  The results of this testing  revealed that NoToX had
antiseptic potency equal to formaldehyde and 5 times greater than  phenol.  In a
comparative Kirby Bauer disc inhibition study also conducted by Earth Safe, the
results found that NoToX had fungicidal activity at least twice that of formaldehyde and
5 times that of phenol.

COST/PAYBACK  OF DEMONSTRATION

      The final cost of the demonstration was $35,653, with $25,000 provided  by EPA
through the Pollution Prevention By and For Small  Business Grant Program.


                                       25

-------
RESULTS AND DISCUSSION

PERFORMANCE RESULTS

Anatomical Performance

      NoToX Biological Preservative was used effectively as an anatomical perfusion
fluid, preserving cadavers as well as formaldehyde and phenol embalming solutions.
Optimal results were obtained when the fluid was perfused through the carotid artery
at 20 pounds of pressure using 8 to 10 gallons of fluid.  Supplemental injections to the
abdominal region enhanced fixation of the viscera.  Cadavers embalmed with NoToX
were firm, yet pliable and had  a lifelike texture that made them well-suited for medical
school, anatomical, and surgical dissection.  Some surgeons preferred NoToX
embalmed specimens for practicing surgical techniques.  A member of the
Arthroscopy Association of North America evaluated knee joint specimens preserved
in NoToX and indicated that these were viable alternatives to fresh frozen specimens
used in teaching laser surgery techniques to orthopedic surgeons.  In addition to
demonstrating long term preservation, NoToX Biological Preservative  also showed its
ability to  inhibit microbial proliferation as an  embalming solution.  Since the two
fundamental criteria of anatomical embalming were met, it was concluded that NoToX
Biological Preservative is an effective substitute for formaldehyde embalming fluids.
Air levels were measured around specimens embalmed in NoToX and there was no
off-gassing of toxic substances. .Students found NoToX embalmed specimens did not
exhibit unpleasant odors.

Histoloqical Performance

      Tissue fixed in  NoToX Histological Fixative was diagnostically equal to tissue
fixed in formaldehyde. There are some minor differences between NoToX and
formaldehyde fixed tissue, but these do not affect its diagnostic utility. NoToX gives a
light tannish appearance to the tissue that some laboratorians may misconstrue as
unfixed specimens. However,  upon touching and evaluating the tissue microscopically
the specimens are indeed firm and preserved.

      Red blood cells can be  seen  in NoToX-fixed tissue,  but they appear differently
than red  blood cells that are seen in formaldehyde-fixed tissue.  In NoToX, the red
blood cells look like flat, two-dimensional, light orange, outlined discs. The change in
appearance  of the red blood cell does not affect cellular architecture,  which is the
crucial factor for medical diagnosis.

      Specimens fixed in NoToX and specimens fixed in formaldehyde were
compared side  by side and judged on the criteria of cellular  architecture, nuclear
detail, cytoplasmic detail, and strength of contrast. The principal investigators at
Mount Sinai  Medical Center determined that the NoToX-fixed specimens were
diagnostically equal to the formaldehyde-fixed specimens.

PRODUCT QUALITY VARIANCE

      Product quality variance is not an issue.  NoToX Biological Preservative and
NoToX Histological Fixative come ready to use and can be made with a high degree
of consistency from lot to lot. Any performance variables are due more to the


                                      26

-------
individual techniques of end users.  If established protocols are followed, the end user
will obtain effective results. For example, Earth Safe recommends that NoToX
Biological Preservative be perfused through the carotid artery at 35 pounds of
pressure and that 8 to 10 gallons of fluid be used for optimal preservation.  NoToX
Histological Fixative is recommended to be used without dilution to achieve optimal
results.  When used according to directions, there is no variation in fixation or
preservation quality.

CONDITIONS THAT IMPACT PERFORMANCE

      NOToX Biological Preservative and NoToX Histological Fixative usually can be
used like formaldehyde.  However, neither NoToX product can be diluted as is
sometimes done with formaldehyde.  Some minor changes in protocols enhance
performance.

Embalming Procedures

      Adequate fluid should be used for perfusing anatomical specimens.   Earth Safe
recommends eight to ten gallons  per cadaver.  The choice of perfusion artery
sometimes affects results.  The most consistent results have been obtained when the
cadaver is perfused through the carotid artery. Supplemental injections to the
abdominal region are helpful in ensuring that viscera reach a firm texture. Alternately,
the viscera can be placed in a pail containing NoToX Biological Preservative after
dissection has begun; viscera will continue to firm up within 24 hours.  A 30-day
curing period is recommended before dissection is begun.

      NoToX Biological Preservative works well with specimens that are stored at
room temperature and under refrigerated and freezing conditions.  In geographic
locations with high humidity, VaporSafe or TankGuard may be used in conjunction
with NoToX Biological Preservative to fortify the inhibition of fungus formation on the
cadavers. This also would be true when formaldehyde is used to embalm  cadavers.

Histoloqical Tissue Processing

      NoToX Histological Fixative is compatible with automated histological tissue
processors and has been approved for use by Miles on their VIP series. Although
Miles is the largest manufacturer of tissue processors, other brands can be found in
hospitals and privately owned laboratories. Brand and age of the processor may
affect the results. Simple adjustments can be made to the settings on the  processor.
Adjustments to the temperature setting, vacuum setting, or time of cycle setting may
be required.  Adjustments to the automatic settings on tissue processors using
formaldehyde also are required at times to prevent over-fixation of specimens.

TABULATION OF DATA

Environmental Demonstration

      Measurements of analytes  that resulted from air sampling during the anatomical
demonstration are shown in Table 1.
                                      27

-------
TABLE 1.     Airborne Concentrations of analytes with NoToX Procedures
DATE
3-23-92
3-23-92
3-23-92
3-26-92
3-26-92
3-26-92
6-2-92
6-2-92
6-2-92
6-9-92
6-9-92
6-9-92
7-6-92
7-6-92
7-6-92
LOCATION
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
embalming lab
gross lab
gross lab
gross lab
ACTIVITY
background
check
background
check
background
check
embalming
embalming
embalming
background
check
background
check
background
check
embalming
embalming
embalming
dissection
dissection
dissection
ANALYTE
formaldehyde
phenol
ethanol
formaldehyde
phenol
ethanol
formaldehyde
phenol
ethanol
formaldehyde
phenol
ethanol
formaldehyde
phenol
ethanol
AIR CONC.
(PPM)
0.6
0.2
0.3
0.3
0.2
13
0.3
0.8
0.6
0.4
0.1
14
0.2
0.1
2
Histoloqical Demonstration

      Fifty one surgical specimens were processed identically in NoToX Histological
Fixative or neutral buffered formalin (NBF). They were fixed for 24 hours, embedded
in paraffin, then cut and stained with hematoxylin and eosin.  Four features were
assessed on histological examination: architecture, nuclear detail, cytoplasmic detail,
and strength of contrast.  Architecture, nuclear detail,  and cytoplasmic detail were
evaluated to determine disease states.  Strength of contrast refers to staining
compatibility for ease in distinguishing cellular components.  Numerical values were
assigned with 3+ being the highest value and 2+ the lowest.  Drs. Linger and Hague
indicated that the numerical values were subjective and that the two fixatives - NBF
and NoToX - were very close, practically indistinguishable. On certain tissue types,
NoToX produced superior results. These were on fatty and muscle  tissue such as
breast and uterine specimens. After completing  the study, Drs. Linger and Hague
                                       28

-------
concluded, "Overall diagnostic quality of tissues was maintained in tissues fixed in
NoToX. NoToX Histological Fixative can give pathologists the quality of sections they
require, as well as a product that is occupationally and environmentally safe. All of the
immunohistochemical studies performed revealed no differences in quality or quantity
of staining of NoToX and formalin fixed tissues."

Microbiological Data

      The objective of this study was to determine the antimicrobial effectiveness of
NoToX Biological Preservative and NoToX Histological Fixative.  Formaldehyde has
antimicrobial properties as well as preservation properties.  Phenol is commonly used
in conjunction with formaldehyde to enhance the antimicrobial effectiveness of
formaldehyde, particularly in the preservation of human and animal cadavers.  To
evaluate NoToX Biological Preservative and NoToX Histological Fixative's  ability to
inhibit pathogens, two studies were conducted: an MIC Study and a Kirby Bauer Disc
Inhibition Study.

      The purpose of the MIC was to show the lowest concentration at which a
preservative would exhibit antimicrobial activity. (See Table 2.)  In this study, NoToX,
formaldehyde, and phenol were diluted until they lost antimicrobial activity (dilutions
below 3 percent were not attempted).  The lower numbers represent greater potency.
For example, NoToX was effective against E. coli when diluted to a concentration of 3
percent of its normal concentration, whereas phenol could only be diluted to 25
percent of its normal concentration before it lost potency.  In the case of A. niger, a
fungus, NoToX and formaldehyde still had potency at 3 percent of their normal
concentrations, whereas phenol was not even effective at the 100 percent level.  Earth
Safe's results from this study found that NoToX has antiseptic potency equal to
formaldehyde and 5 times greater than that of phenol.

TABLE 2.     MIC Values (%) for Bacteria and Fungi
Preservativ
e
NoToX
10%
Formalin
5% Phenol
CaroSafef
E.
coli
3
3
25
Nl
B.
meg
3
3
Nl*
Nl
P.
aerug
3
3
12
Nl
S.
aureus
3
3
50
Nl
A. niger
3
3
100
Nl
R.
stolonifer
3
3
25
Nl
* No inhibition
t Preservative manufactured by Carolina Biological

      The Kirby Bauer Study was conducted to compare antifungal potency for
various preservatives as shown in Table 3. In this study, paper discs were
impregnated with the preservatives and placed in a nutrient broth that would grow the
organism of interest. The numbers reported in Table 3 indicate diameters of zones in
which a microorganism could not grow.  With respect to A. niger, the inhibition zone
for NoToX was 60 millimeters, that for formaldehyde was 32 millimeters, and that for
                                       29

-------
phenol was 8 millimeters. This indicates that NoToX has superior potency against the
fungal organism A. niger. For the organism R. stolinifer,  NoToX showed the inhibition
zone of 38 millimeters while formaldehyde and phenol showed 0 inhibition zones.  The
interpretation of the disc sizes indicates that NoToX has fungicidal activity at least
twice that of formaldehyde and 5 times that of phenol.

TABLE 3.   Size of Disc Inhibition Radii (mm) Relating to Kirby Bauer Anti-fungal  Disc
            Inhibition Study
Preservative
NoToX
10% Formalin
5% Phenol
A. niger
60
32
8
B, stolonifer
38
0
0
COST/BENEFIT ANALYSIS

      Formaldehyde has been in use for over 130 years and is a relatively
inexpensive fixative for initial costs.  However, in today's regulatory and environmental
climate many indirect costs impact formaldehyde use.  These include the costs of
hazardous waste disposal, training and safety programs for employees exposed to
formaldehyde vapors, administrative costs to meet OSHA and EPA standards,
monitoring expenses, safety gear, installation and  maintenance of engineering
controls, and the liability of Medical Removal Protection for each employee exposed to
formaldehyde. The potential liability also exists for OSHA fines when permissible  air
levels go above the 0.75 ppm  standard. These fines have ranged from $5,000 to over
$40,000 per  incident.

      All of these liabilities can be eliminated when NoToX is adopted in the
laboratory.  If the purchasing agent considers the  indirect costs of formaldehyde use
and compares them to the direct cost of the NoToX products, she will find that NoToX
is a highly cost effective measure.  Table 4 shows a sample waste disposal  cost
comparison of NoToX Histological Fixative and formaldehyde for a histological
application.

TABLE 4.    Waste Disposal Cost Comparison: Histological Application of NoToX
            Versus Formaldehyde
Product
Formaldehyde
NoToX H.F.
Savings derived whc
gallon
Ave. Direct Cost
(per gallon)
$10
$17
Ave, Disposal
Co$t$ (per
gallon)*
$21
$0
Total Cost
(per gallon)
$31
$17
3n NoToX is used 	 $14 per

                                      30

-------
*assuming< a hazardous waste disposal cost of $1,300 per 55-gallon drum, based on
a quote from S&W Waste, NJ; 50 gallons of a 55 gallon drum of formaldehyde would
be wasted and disposed as hazardous


      When these disposal costs are factored into the direct cost of formaldehyde
(which averages about $10 per gallon) the actual costs of using formaldehyde rises to
$31 per gallon or more.  The average cost of NoToX is $17 per gallon, which  means
there is an actual savings of $14 per gallon.

      Considering that there are 7,000 hospitals and private laboratories in the U.S.
using as much as 5 million gallons of formaldehyde annually, the total savings that can
be derived from using a formaldehyde replacement such as NoToX Histological
Fixative could be significant.

CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      The testing and research done on NoToX Biological Preservative and NoToX
Histological  Fixative during the project period resulted in a technology that was
optimized for market acceptance.  The regulatory climate surrounding formaldehyde
may accelerate the acceptance of these products into the marketplace.  (For  example,
OSHA lowered the permissible levels of formaldehyde exposure to 0.75 parts per
million (ppm), further encouraging laboratories to at least evaluate NoToX Biological
Preservative and NoToX Histological Fixative.)  Finding a formaldehyde replacement
that would perform all tasks traditionally delegated to formaldehyde has been a major
problem to those who wanted to discontinue formaldehyde use.  NoToX  is filling that
need in the  marketplace and is being distributed worldwide and gaining acceptance.

       NoToX Biological Preservative and NoToX Histological Fixative reached their
final phase of development, and no further developmental work is planned. Some
product improvements may be considered in the future,  but the products are now
performing to meet  the needs and expectations of the marketplace.

       NoToX Biological Preservative is targeted for several markets. These include
anatomy laboratories at medical, dental, and veterinary schools; the funeral industry;
companies dealing  in artificial organ components; and companies dealing with dialysis
supplies.

       NoToX Histological Fixative is targeted for hospital and  private medical
laboratories as well  as research institutions.  The Histological Fixative also is being
used as an  environmentally safe fixative for parasitology and is being distributed by
Scientific Device Lab, Chicago, Illinois, which caters to the parasitology and
microbiology market.

       Another event that enhanced the credibility of NoToX Biological Preservative
and NoToX Histological Fixative was the R&D 100 award that designated NoToX as
one of the 100 most technologically significant products of the year. NoToX Biological


                                       31

-------
Preservative and NoToX Histplogical Fixative are now being marketed worldwide.  The
study provided the opportunity to fine tune the formulas to ensure they would meet
the needs of end users.

Barriers

      There appears to be no technical shortcomings.  Pricing has been somewhat of
an issue since the customer is accustomed  to paying a  nominal price for
formaldehyde. However, as the customer becomes educated about the true costs of
formaldehyde use, the price of NoToX becomes less of a barrier.

      The solution to overcoming the pricing barrier is to educate the consumer by
meeting them at national and regional association meetings.  Salespeople are making
direct contact with these consumers,  and telemarketing,  advertising,  and direct mailing
efforts are used to highlight the cost effectiveness of NoToX. Another solution to the
pricing barrier is manufacturing the product  more cost effectively which will occur as
the volume of demand increases.
                                     32

-------
    SUBSTITUTION OF BIODEGRADABLE LOW TOXICITY NATURAL PRODUCTS
                       FOR THE KILLING OF FIRE ANTS

                                      by

                               Joe S. Wilkins, Jr.
                                Joe Wilkins, Sr.
                         Environmental Pesticides Group
                             Pasadena, TX  77502


ABSTRACT

      Environmental Pesticides Group found that a low toxicity, biodegradable, natural
products (terpines isolated from citrus fruit peel) can be substituted for toxic chemicals
for the effective killing of fire ants in agricultural settings or on residential lawns. The
fire ant has invaded the southern United States causing great damage to farm animal
production, agricultural product harvesting, native ant populations, and recreational
lands. This project showed that citrus fruit terpene compounds are (1) as effective as
commonly used insecticides in killing fire ants; (2) less expensive than commonly used
toxic chemicals; and (3) rapidly biodegradable in the environment.


INTRODUCTION

PROJECT DESCRIPTION

      The fire ant,  Solenopsis invicta, a native of South America, was inadvertently
introduced into the United States at the port of Mobile, Alabama in the 1930s.  Since
that time, it has spread throughout the southeastern and southcentral U.S. where it
currently occupies over 250 million acres, spanning all or parts of 11 states and
Puerto Rico.  Its aggressive nature, potent sting, large colony size (upwards of
250,000 workers  per nest), and feeding and mound building habits make this ant a
particularly noxious pest.  The problems caused by this species are numerous and
include stinging people and their animals, harming wildlife, defacing lawns and parks,
and interfering with farming operations.  The problem is growing worse as the ant
continues to expand its range northward and westward. Should the ant become
established on the  West Coast, it is expected to thrive throughout coastal California,
Oregon, and Washington.

      In Texas, about half the state -- 65 million acres -- is infested and under a
United States Department of Agriculture (USDA) quarantine restricting the movement
of many products and materials from these  areas to uninfested areas.  The Texas
Department of Agriculture has estimated that the fire ant costs the state at least $31
million per year in losses and damages. In  the infested areas, farmers, commercial
turf and plant producers, and home owners are engaged in a constant battle to
protect their products and  homes. In order to fight this battle they are using a wide
spectrum of toxic chemicals on a  routine basis. The Texas Department of Agriculture
has published a list of over 100 products registered for fire ant control; all are  toxic to
humans and other  mammals and  pose some threat to the environment.
                                      33

-------
       The most popular form of fire ant control remains the spraying of toxic
 chemicals by agricultural concerns, residential pesticide companies, and home
 owners. These chemicals and their by-products contribute significantly to residuals
 found in public wastewater treatment plants, public water supplies, and residential
 yards.  Development of a non-toxic, cost effective  method of fire ant control would
 greatly reduce pollution (through source reduction) caused by the thousands of small
 business agricultural sprayers, residential exterminators, and home owners in the fight
 against this noxious pest.

       The Environmental Pesticides Group is perfecting a non-toxic, unique fire ant
 pesticide that is highly effective in controlling S.  invicta and replaces the toxic
 substances currently in use.  The product is a blend of two environmentally safe
 ingredients with water.  The active ingredient, d-limonene, is a natural terpene product
 isolated from citrus fruit peel. D-limonene is blended with Mazclean,  an emulsifying
 agent, and the resulting mixture is diluted with water to achieve a solution containing 3
 to 5 percent d-limonene. This final formulation,  called Dead Ant, has no known
 toxicity, flammability, or environmentally hazardous effects.  Environmental Pesticides
 has found that Dead Ant can be as effective as  commonly used toxic insecticides in
 killing fire ants in both the laboratory and a commercial field setting.  In addition, the
 active ingredient, d-limonene, degrades rapidly,  almost disappearing from the soil in
 four days.

 Unique Product Features

       The unique  properties  of Dead Ant come from the unusual way in which it
 appears to affect the target pest.  Although not  scientifically documented, the
 observations of Environmental Pesticides Group suggest that d-limonene destroys  the
 outer protective layer of the ant cuticle resulting  in death.  Dead Ant works on contact
 and permeates the soil quickly,  rapidly killing the workers, larvae,  and queens. The
 product's major advantage  is that it is non-toxic and does not introduce harmful
 chemicals into sensitive areas (e.g., yards,  playgrounds, workplaces, barns, and the
 like).  D-limonene is listed by  the United States Food and Drug  Association (FDA) and
 Federal Emergency Management Association (FEMA) as GRAS (generally recognized
 as safe), permitting its use in  cosmetics, and in the food industry as an additive for
 flavoring and coloring foods and beverages. The emulsifier, Mazclean EP, is not
 considered a hazardous chemical or a priority pollutant by the manufacturer's (PPG
 Industries) Material Safety Data Sheet.

 Process Schematic

      The product preparation is shown in  Figure 1. First d-limonene is blended with
the emulsifying agent, Mazclean. The resulting mixture is blended with water in a large
blending tank.  The final solution, containing 3 to 5 percent d-limonene, is then
pumped into smaller containers.
                                       34

-------
Figure 1.    Dead Ant Product Preparation
                           PUMP
 APPLICATION

 Products Replaced

       The Environmental Pesticides Group found that a class of bw toxicity,
 biodegradable, natural products -- terpines isolated from citrus fruit peel --  can be
 may substitute for toxic chemicals such as diazanon, dursban, malathion, arsenic,
 bromide gas, and trichloroethane for effective control of fire ants.  Use of natural
 terpenes for fire ant control is a novel concept; Environmental Pesticides Group was
 unable to find published records of the application or the effectiveness of d-limonene
 for killing fire ants.  The retail cost of Dead Ant is expected to be comparable or less
 expensive than other commonly used toxic products.

 Wastes Prevented

       The major waste prevention afforded by Dead Ant is its lack of environmental
 contamination. The product will not contribute to residuals found in public  wastewater
 plants, public water supplies, and residential yards because it is non-toxic and rapidly
 degrades in the environment.  The precise cost of residual pollution due to the current
 fire ant control products is difficult to estimate. However, the reduced risks to human
 health and the environment is expected to be considerable if Dead Ant was to be used
 extensively  in place of other currently available products.
                                       35

-------
 Cross Segment Uses

       In addition to the control of fire ants, Dead Ant has excellent potential for
 control of other pest ants such as the Pharaoh's ant, Monomohum pharaonis  and the
 carpenter ant, Camponotus pennsylvanicus. Other insect pests may also be
 controlled with the product, and Environmental Pesticides Group is planninq future
 tests with d-limonene.
 PROCEDURE
       Environmental Pesticides' project consisted of three studies: (1) laboratory
 studies of fire ant mortality; (2) field studies of fire ant mortality; and (3) product
 degradation under field conditions.

 DEMONSTRATION  PROCEDURES

 University of Texas  Laboratory Studies of Dead Ant Efficacy

       Laboratory tests were conducted by Dr. Edward Vargo at the University of
 Texas at Austin.  Source colonies for this study originated from the Brackenridge Field
 Laboratory of the University of Texas at Austin and were collected on 23 April 1992
 Tests were conducted 26 April-21 May 1992. All colonies  were of the multiple queen
 (polygyne) form.  Small  standardized treatment colonies were constructed consisting
 of 10 grams of workers  and brood (all immature stages of fire ants) -- approximately
 5,000 workers, 5,000 worker pupae, and 5,000 larvae - and 5 queens  The ants were
 housed in plastic trays (40 x 27 x 9 centimeters with  Fluon®-coated sides to prevent
 escape), each containing a nest (a 14 centimeter diameter Petri dish half-filled with
 moist plaster).  The ants were able to move around in the  trays.  The test colonies
 were kept at room temperature, approximately 75°F.  There were 5 replicate test
 colonies per treatment.  Each treated colony received approximately 0.5 fluid ounces
 of product delivered as a fine mist with a hand held spray  bottle. This method of
 application was enough to ensure that almost every exposed ant was coated with
 material.

      Two methods of application were tested, simulating  mound drench and surface
 spraying.  Simulation of mound drench was performed by treating the colonies with
 the nest cover removed  so that all ants, including brood and queens, were exposed
 and sprayed directly with the solution.  Simulation of surface spray of the  mound was
 achieved by treating  the colonies with the nest covers on, thereby shielding those ants
 -- brood^ queens, and approximately 80 percent of the workers -- that would normally
 remain below ground, unexposed to direct spray.  Four concentrations of d-limonene
 solution were tested:

      1.5 percent: 1.5 percent d-limonene, 0.75 percent Mazclean emulsifier
      3.0 percent: 3.0 percent d-limonene, 1.5 percent  Mazclean emulsifier
      5.0  percent: 5.0 percent d-limpnene, 2.5 percent  Mazclean emulsifier
      10.0 percent:  10.0 percent d-limonene, 5.0 percent Mazclean emulsifier

      The control consisted of untreated colonies. A 10 percent solution  of the
Mazclean emulsifier with  no d-limonene was also tested.
                                      36

-------
      Efficacy of the treatments was determined 24 hours, 1 week, and 3 weeks after
treatment by weighing all surviving ants to determine the effect on workers and brood,
and counting the number of living queens to determine queen mortality.  Monitoring
the surface spray simulation (nest cover on) was ended after 24 hours, at which time it
was determined that this method was not as effective as the drench simulation. (See
Figure 2 and Tables 1 and 2.)  Statistical comparisons among treatments were made
by analysis of variance tests and the Newman-Keuls  multiple range test.

Field Studies of Dead Ant Efficacy

      There were two tests of the effectiveness of Dead Ant in the field:  a test by the
Environmental Pesticides Group and one by personnel at Texas A & M University.

Environmental Pesticides Group Field Test

      Preliminary tests were performed at Mr. Bill Murff's grass farm in Crosby, Texas
to evaluate the effectiveness of various concentrations of d-limonene solutions in
controlling fire ants  in field situations.  Four test plots were demarcated, and all
mounds within the test plot were flagged. In  addition, all mounds within  6 feet of the
perimeter of each plot were also marked so that any movement of treated mounds
outside the test plot could be detected.  Mounds in each plot were treated with a
single concentration of d-limonene solution by spraying an even coat of  material using
a hand-held sprayer.  In the larger plots  (40 x 50 feet), approximately 2 gallons of
solution was sprayed; approximately 1 gallon of solution was sprayed on the smaller
plots (10 x 20 feet).  After spraying, mounds were treated by drenching with 2 gallons
of solution.  The size of the plots, the number of mounds in the plot, and the
concentration of applied solution was as follows:

       1.5 percent (+ 0.75 percent Mazclean): 10 x 20 foot plot contained 2 mounds

       3.0 percent (+ 1.5 percent Mazclean): 40 x 50 foot plot contained 7 mounds

       5.0 percent (+ 2.5 percent Mazclean): 40 x 50 foot plot contained 6 mounds

       10.0 percent (+ 5.0 percent Mazclean): 10 x 20 foot plot contained 4 mounds

       Treatments were applied on 15 April 1992. These test plots were also used as
the source for the product degradation studies.

Texas A & M University Field  Test

       The test was conducted by Dr. Bastian M. Drees and Mr. Charles L. Barr of the
Texas A & M Agricultural Extension Service.  The test began on 5 August 1992 on the
earthen dam at Lake Somerville, Texas.  Plots of land 40 feet wide and of variable
length were marked such that each  plot contained 10 active fire ant mounds.   The
plots were grouped into 3 blocks (replication) containing 6 plots each, and treatments
were randomly assigned within the blocks.  Mound density on the dam was
approximately 380  mounds per acre, a density indicative of a population of multiple-
queen  colonies.  Mound activity  was determined by  lightly disturbing the mound with a
 pointed wooden tool handle.  A  mound was  considered active if within 15 seconds of
 disturbance a number of ants came to the surface.


                                       37

-------
       Treatments were applied beginning at 9:00 am and ending at 4:00 pm. The
 applied solution was mixed in a large drum and dispensed into 2-gallon sprinkler
 cans.  The diffusion nozzle was removed from the end of the spout to give a solid
 stream permitting better penetration of the mound surface. The following treatments
 were applied:

 •     1.5 gallons per mound of 3 percent d-limonene (with 1 percent Mazclean)
       1.5 gallons per mound of 4 percent d-limonene (with 1 percent Mazclean)
       1.5 gallons per mound of 5 percent d-limonene (with 2 percent Mazclean)
       1.5 gallons per mound of 2 teaspoons Orthene® per gallon of water
 •     1.5 gallons per mound of water as a control

       Upon mixing with the water, the d-limonene solutions became unusually
 viscous.  This may have been because the water had been sitting in a tank for some
 period and contained a considerable amount of algae.  The thickness of the material
 reduced its penetration into the mounds, causing it to pool on top of the mounds for
 some time rather than immediately percolating down. The material did eventually seep
 down into the mounds.

       Post-treatment evaluations began at 9:00 am on 6 August (1 day post-
 treatment), 7 August  (2 days post-treatment),  and 12 August (7 days post-treatment)
 The data were analyzed using an analysis of variance test, and the Newman-Keuls
 multiple range test was employed to determine significance of differences amonq
 treatments (P<0.05).

 Product Degradation

       Mercury Environmental Services, Pasadena, Texas analyzed the soil samples
 taken from the test plots on Mr. Bill Murff's grass farm in Crosby, Texas using  gas
 chromatography/mass spectrometry (GC/MS) equipped with a purge trap.

       Five point calibrations of d-limonene were run with concentrations at 20 40  60
 80, and 100 parts per billion. The GC/MS parameters were 40°C to 220°C with an  '
 initial hold time at 2 minutes, ramping at 8°C per minute, with a final hold time of 4
 minutes.  With these parameters, five sets of samples were analyzed from the four test
 plots.

      Samples were  taken from random locations within the plot, but at least 6 feet
from the nearest drenched mound.  One sample per plot was collected per sample
period. The sample plugs were 4.5 x 4.5 x 6 inches deep. Upon collection samples
were placed directly into 1 quart glass jars for transport to the laboratory for analysis
      The cost of the demonstration project was $35,401, with $25,000 provided by
EPA through the  Pollution Prevention By and For Small Business Grant Program and
$10,401  provided by the Environmental Pesticides Group
                                     38

-------
RESULTS AND DISCUSSION

PERFORMANCE RESULTS

University of Texas Laboratory Evaluation of Dead Ant Efficacy

      There were highly significant differences in mortality among treatments at all
three time periods for both workers/brood (day 1: F™  = 65.4,^P^< 0.0001; days 7
and  21:  F529 > 98.4, P < 0.0001; and queens (day 1:7^ = 41.0, P  < 0.0001; days
7 and 2V FL > 26.2, P < 0.0001) (see Figure 1 and Tables 1 and 2).  Mortality of
both workers/brood and queens in the surface spray simulation treatments after 24
hours was not significantly different from the untreated colonies (Figure 3). However,
application of the d-limonene solution in the drench simulation, at all concentrations
tested resulted in significant mortality of workers/brood after only 24 hours. There
was an obvious concentration effect at the day 1 and day 7 periods, with the 1.5
percent solution being significantly less effective than the three higher concentrations
tested and the 5 percent  and 10 percent solutions killing all individuals in the colonies
within 24 hours  By day 21, this concentration effect disappeared, and the four d-
limonene concentrations did not differ statistically in percent mortality.  The Mazclean
control also resulted in significant mortality at all time periods.  Although there was
significantly less mortality  than in any of the d-limonene solutions at the day 1  and day
7 periods on day 21  mortality in the Mazclean-only treatment did not differ significantly
from any of the d-limonene treatments, at which time all treatments resulted in over 90
percent mortality.  This result suggests that the high concentrations of Mazclean (10
percent) used in the Mazclean-only treatment is toxic to the workers/brood, but that
this toxicity is delayed compared to the d-limonene solution.
                                        39

-------
Figure 2.    Effect of D-limonene Solutions on Fire Ant Mortality in Laboratory
            Colonies.  Each colony started with 10 grams of workers and brood (i e
            eggs, larvae, and pupae) and 5 queens. Mortality of workers and brood'
            is shown separately from that of queens for each treatment on each of
            three evaluation periods.  There were 5 replicates of each treatment.
            Shown are means ± SD. Treatments with different letters on a particular
            day differed significantly (P  <.0.05, Newman-Keuls multiple range test)
            Although only the treatments receiving the drench simulation are shown
            in this Figure, the analysis of variance tests and multiple range tests
            included the treatments receiving the  surface spray simulation on day 1
            (see Figure 3). These data are also presented in Tables 1 and 2 (see
            section "Tabulation of Data").
                  Day 1
Day 7
Day 21
                                     40

-------
      Differences among treatments were more pronounced when queen mortality
was considered (see Figure 2 and Tables 1 and 2). At all three time periods there was
significantly higher mortality in the three highest concentrations of d-limonene than in
all other treatments.  As with worker/brood mortality, the 5 percent and 10 percent
solutions resulted in 100 percent  mortality.  By  day 21, mortality in the 3 percent d-
limonene solution also reached 100 percent.  There was no queen mortality in the
Mazclean-only treatment by day 7, and only slight mortality (16 percent) by day 21, but
this was not significantly more than occurred in the untreated control.


Figure 3.     Effect of Spray Simulation Application of D-limonene on  Laboratory
             Colonies of Fire Ants 24 Hours After Treatment.  Colonies (n = 5 for each
             treatment) had the same composition as those reported in Figure 1.
             Treatments with different letters differed statistically  P < 0.05,  Newman-
             Keuls test).  In the statistical analyses, the above treatments were
             combined with the drench simulation shown in Figure 2.  These data are
             also presented in Tables 1  and 2.
                             Percent mortality             Percent mortalrty
                                           .4
                          -*ioo>»wg>2j9>$S
                         ooooooooooo
                          ! I I • I « I « I ' I

             Mazclean control | 0




             1.5% d-llmonene



               3% d-limonene



               5% d-limonene
              10% d-limonene
                                         41

-------
 Conclusions

       The d-limonene solutions effectively killed the ants within 24 hours after
 treatment when the liquid was directly applied to all ants through the drench
 simulation method.  The fact that there was only limited mortality in the surface spray
 simulation application indicates that the material must come into direct contact with the
 ants to be effective, and that there is no appreciable transfer of material among ants
 through normal social interactions. When applied directly to the ants, there was a
 concentration effect with increasing mortality from solutions of 1.5 percent to 3 0
 percent to 5.0 percent. There was no difference between the 5.0 percent and 100
 percent formulations, as both treatments resulted in 100 percent mortality of
 workers/brood and queens.  The measure of queen mortality is the most important
 because without a queen the colony will die, whereas as long as there is at least one
 active queen, she can lay eggs to replace any workers or larvae that might be killed
 The Mazclean emulsifier had a delayed effect primarily on  the workers and brood  with
 little effect on the queens.  The substantial  difference between the efficacy of the d-
 limonene treatments and the Mazclean-only treatment shows that is the d-limonene
 and not the emulsifier, that is  responsible for the main effects of the formulation
 Environmental Pesticides finds additional support for this conclusion in that the
 concentration of emulsifier used in the Mazclean-only treatment (10 percent) was twice
 that of the highest concentration than in any of the d-limonene formulations tested
 (maximum concentration of Mazclean was 5.0 percent).

 Field Test of Dead Ant Efficacy

 Environmental Pesticides Group Field Test

       Upon thorough examination of all treated mounds on 21 April, 6 days after
 treatment, no fire ant activity could be detected (Figures 4 through  7).  Moreover no
 new mounds (i.e., previously unmarked mounds) were observed in or  near the  '
 treatment plots. Although no formal control mounds were used in this test several
 untreated mounds around the perimeter of the test plots showed vigorous signs of fire
 ant activity upon  inspection. All concentrations of the d-limonene solutions aooeared
 to totally eliminate the treated  colonies.

 Texas A & M University Field Test

       A significant difference was noted among treatments at all three post-treatment
 test periods (F512, > 4.92, P < 0.02)  (see Figure 8 and Table 3).  At  day 1  all
 treatments containing active ingredient (i.e., all the d-limonene solutions and Orthene®
 a class III insecticide known for giving fast effective results  in controlling fire ant
 mounds) rendered a significant percentage of mounds inactive.  The 5 percent d-
 limonene was the most effective of all treatments with 83 percent reduction in the
 number of active  mounds, but this was not significantly more than the other active
 ingredient treatments. At day 2, all active ingredient treatments still  showed
 significantly more activity than  the water control, but activity was also evident in the
 Mazclean-only treatment, suggesting delayed action of this material   Only the 5
 percent d-limonene and Orthene treatments - which gave 80 percent and 90 percent
 effectiveness  respectively - showed significantly more activity than  the Mazclean-onlv
treatment.  The delayed action of Mazclean was more evident than at day 7  at which
                                       42

-------
time it showed the third highest activity only behind the 5 percent d-limonene and the
Orthene® treatments, although it did not differ significantly from these treatments.

Conclusions

      All d-limonene solutions, especially the 5 percent solution, compared favorable
with Orthene®  The d-limonene treatments may have been  even more effective had
the applied material not been so viscous. As with the laboratory studies, there was a
delayed but significant effect of the Mazclean applied by itself. However, this effect
cannot account for the quick destruction of the fire ant mounds observed in all of the
d-limonene treatments. It is the d-limonene that is responsible for the main effects
seen in those treatments.  These results show that the mortality observed in the
laboratory after treatment with d-limonene can be produced in field colonies through
mound drenches.
                                        43

-------
Figure 4.     Test Plots Layout and Application Results of  1.5 Percent D-limonene to
               Fire Ant Mounds at a Grass Farm
                   DATE  fr-21-92
                                      TIME  11:15 an
                                                          WEATHER Clear--78
                   LOCATION
                            PLOT
                                          -SIZE.
                                                10x20
                   FORMULATION  X    I 1/2S S-1QOI.S
























m
r
i
L
•t^



















L
1

A
.^^i

























^
— 1
\





















M
•
•
F




















L
m~
T_
r-














































































































NO. OF 2-Beds- 100% Kl< L*-NO
BEDS ^ign gf jftivitv.




SIZE/BEDS SEE lt-1^-92


DATES '•-15-92
PI line U" \ A-Q?
TAKEN *- 17-92
(•-19-92
l»-2l-92

DATES '•-15-92
CPPA vrn


DATES '•-15-92
DRENCHED


AHOUNT SEE '•-1 5-92
4v ncr
DRENCH

              N°t«:  HeavX r'in at  '0=00 p«- 4-15-92- I" plus




              NOTED:  Mo n«w beds within or outside Plot

                     Light to no yellow effect to grass around drenched beds.
                                              44

-------
Figure 5.     Test Plot Layout and Application Results of 3.0 Percent D-limonene to
               Fire Ant Mounds at a Grass Farm
                                                             WEATHER Clear—78 degrees
                   Note  Heavy ram at 10 00 pm-  I»-I5'32- I"
                   NOTED:  No new beds within or outside Plot
                         Light yellow grass around drenched beds-new growth seen. Grass not  dead.
                          Bed 5- Outside Plot was treated by Probe  (2V) deep on k sides.
                               the effect was only 75% Kill
                                                  45

-------
Figure 6.     Test Plot Layout and Application Results of 5J3 Percent D-limonene to
               Fire Ant  Mounds at a Grass Farm
                   DATE
                                       TIME
                   LOCATION PLOT Y
                                           SIZE
                   FORHULATION X   5^5-1005
                                                           WEATHER Clear--78 degrees
                                                        NO.  OF    6-8eds-10Q% KIU--No
                                                        BEDS      sZZHHIIIIZZI
                                                        SIZE/BEDS 5rg.  JI.IC.Q->
                                                       DATES
                                                       PLUGS
                                                       TAKEN
                                                       DATES
                                                       SPRAYED
                                                       DATES
                                                       DRENCHED
                                                                 (,-15-92
                                                                 4-10-q?
                                                       AHOUNT    SEE it-15-92
                                                       t X OF    	
                                                       DRENCH    	
              NOTE:  Heavy r«jn at 1Q:QO pm-  ^-l«;-32- 1" plus

                                                                          I
              NOTED:  No new beds within or outside Plot

                    Light yellow effect on  grass around drenched beds. Grass not dead
                                                 46

-------
Figure 7.     Test Plot Layout and Application Results of 10.0 Percent D-limonene to
               Fire Ant Mounds at a Grass Farm
                   DATE
                                      TIKE
                                             10:50 am
                   LOCATION
                            PLOT W
                                          SI2E
                                               10x20
HEATHER Clear-- 78 degrees
                                                       NO. OF
                                                       BEDS
                                                       SIZE/BEDS  SEE
                                                       DATES
                                                       PLUGS
                                                       TAKEN
                                                                  U-21-92
                                                       DATES      tt-!S-Q2
                                                       SPRAYED    	
                                                        DATES      4-15-92
                                                        DRENCHED   	
                                                        AHOUNT
                                                        I X OF
                                                        DRENCH
                                                                  SEE »- 15-92
               Note:  Heavy rain at.10:00 pm- U-lf-92- 1" plus


               NOTED:  No new beds within or outside Plot
                      Grass around drenched beds are yellow, not dead.
                                                  47

-------
Figure 8.    Field Tests of Efficacy of D-limonene to Control Fire Ant Mounds.  Shown
            is the percent (± SD) mounds out of 10 treated in each plot that were
            inactive when examined. Three replicates (plots) were made of each
            treatment.  Treatments with different letters on  a particular day differed
            significantly (P < 0.05, Newman-Keuls multiple range test).  The data are
            also presented in Table 3.
                      1 day


0)
•o
0
0)
is
c
"c
Q.


100
90
80
70
60
50
40
30
20
10
0
                              2 days
           4>
           «5
                *
                o
                     CO
                           •o

                           Sr
                                m
                                     o>
                                     -C

                                     O
                                                  "
                                         O

                                         •6
                                         ss
                                         in
o>
x:
•c
O
                                    7 days
                          0)
                          "co
1
o
                                                    •c
                                                    o
                                    CO
                                               IO
                                      48

-------
Product Degradation

      In most plots, there was an initial increase in the amount of d-limonene in the
soil samples for the first 2 days after application (see Figure 9 and Table 4), but the
product rapidly decreased after this period. By day 6, the quantity of d-limonene had
fallen to 0.008 milligrams per kilogram of soil or less in all plots. The reason for the
high quantity of d-limonene present of day 1 in the 10 percent plot is not known;  even
in this case, there was rapid degradation of the product by day 4.

Overall Conclusions

      Both in the laboratory and in the field, the 5.0 percent formulation of Dead  Ant
proved fast and effective in killing fire ants, including workers, larvae, and queens. In
the field, its performance was equal to that of Orthene®.  As expected, d-limonene
rapidly degraded  in the environment providing fast, effective; environmentally safe
control of fire ants with no long-term residues remaining in the environment after
application.

PRODUCT QUALITY VARIANCE

      Percentages of d-limonene, emulsifier, and water are  very tightly controlled and
will comply with the Texas Department of Agriculture and EPA requirements when
these are determined. Tolerances are kept within 0.01 percent. Judging  from the
viscous product that was obtained when mixed with the sitting water in the Texas A &
M University test,  the purity  of the water is important in producing a final product of
the proper viscosity.  Nevertheless, the thick consistency of the material did not seem
to adversely affect the performance of Dead Ant during this test.

CONDITIONS THAT IMPACT PERFORMANCE

      Although not specifically investigated, environmental conditions  are expected to
affect the performance of Dead Ant. Because the product  must come into contact
with the ants to be effective, the closer the ants are to the surface of the mound,  the
greater the chances of product contacting all colony members when the mound  is
drenched.  These conditions are found in early afternoon in spring and fall when  the
ants come near the soil surface to take advantage of the sun-warmed environment of
the mound. This  may account for the difference in the effectiveness of the field trials
done by Environmental Pesticides Group (performed  in April when high temperatures
were about 75°F)  in which 100 percent control was obtained, and those performed by
Texas A & M (done in August when high temperatures were about 95°F) in which 5
percent Dead Ant gave about 95 percent control. However, control was effective even
in the heat of summer, and  excellent results can  be expected under most
environmental conditions.
                                       49

-------
Figure 9.    Degradation of D-limqnene Under Field Conditions. Samples were taken
            from plots receiving different concentrations of d-limonene sprayed on
            surface of study plots. The samples were analyzed by GC/MS.  The
            data are also presented in Table 4.
                                           1.5%
                                         d-limonene


— _
1
O)
^
0
c
c
O
E

•6
O)
E




4
3
2

1

0
3%
d-limonene





, <• . m . .


5

4

3
2
1
0'
5%
d-limonene




S—m**JSLm m •_• ^_ «~«Jl^9M^bB^HM^I^M^J
                            1   23456

                              Time (days)
                                     50

-------
TABULATION OF DATA

University of Texas Laboratory Evaluation of Dead Ant Efficacy


TABLE 1.    Effect of D-Limonene Solutions on Worker/Brood Mortality in Laboratory
            Colonies.
            Each colony started with 10 grams of workers and brood.  There were 5
            replicates of each treatment. Treatments with different letters on a
            particular day differed significantly (P <  0.05, Newman-Keuls multiple
            range test).

Treatment
Untreated Control
Percent Mortality (mean ± SD)
1 day
4.6 ± 2.3
D
7 days
16.4 ± 3.5
D
Drench Simulation
Mazclean only
1 .5% d-limonene
3.0% d-limonene
5.0% d-limonene
1 0.0% d-limonene
27.2 ± 2.4
C
48.2 ± 34.7 B
99.4 ± 0.9 A
100 ± 0
A
100 ± 0
A
65.4 ± 8.8
C
83.8 ±14.0 B
99.6 ± 0.6 A
100 ± 0
A
100 ± 0
A
21 days
48.8 ± 3.6 B

93.6 ± 8.7 A
95.4 ± 5.9 A
100 ± 0
A
100 ± 0
A
100 ± 0
A
Surface Spray Simulation
Mazclean only
1 .5% d-limonene
3.0% d-limonene
5.0% d-limonene
10.0% d-limonene
17.6 ± 4.5
CD
12.2 ± 6.0
CD
2.6 ± 2.6 D
16.8 ± 3.8
CD
13.0 ± 3.9
CD










                                       51

-------
TABLE 2.    Effect of D-Limonene Solutions on Queen Mortality in Laboratory
            Colonies.
            Each colony started with 5 queens.  There were 5 replicates of each
            treatment.  Treatments with different letters on a particular day differed
            significantly (P < 0.05, Newman-Keuls multiple range test).
Treatment
Untreated Control
Percent Mortality (mean ± SD)
1 day
0
c
7 days
4.0 ± 8.9
B
21 days
8.0 ± 17.9
B
Drench Simulation
Mazclean only
1 .5% d-limonene
3.0% d-limonene
5,,0% d-limonene
10.0% d-limonene
0
c
24.0 ± 32.9
C
72.0 ± 30.0
B
100 ± 0
A
100 ± 0
A
0
B
24.0 ± 43.4
B
88.0 ± 26.8
A
100 ± 0
A
100 ± 0
A
16.0 ± 26.0
B
36.0 ± 35.8
B
100 ± 0
A
100 ± 0
A
100 ± 0
A
Surface Spray Simulation
Mlazclean only
1 .5% d-limonene
3.0% d-limonene
5.0% d-limonene
1 0.0% d-limonene
0
C
8.0 ± 11.0
C
4.0 ± 8.9
C
4.0 ± 8.9
C
0
C










                                      52

-------
Field Tests of Dead Ant Efficacy:  Texas A & M University Field Test
TABLE 3.    Field Tests on the Efficency of D-Liomene Solutions to Control Fire Ant
            Mounds.                                                ,   ,
            Shown is the percent (± SD) mounds out of 10 treated in each plot that
            were inactive when examined. There were 3 replicates (plots) of each
            treatment.  Treatments with different letters on a particular day differed
            significantly (P < 0.05, Newman-Keuls multiple range test).
Treatment
Water control
Mazclean only
3.0% d-limonene
4.0% d-limonene
5.0% d-limonene
Orthene®
Percent Mounds Inactive
1 day
3.3 ± 5.8
B
10.0 ± 0.0
B
66.7 ± 25.0
A
73.0 ± 15.0
A
83.0 ± 12.0
A
73.0 ± 15.0
A
2 days
6.7 ± 5.8
C
40.0 ± 17.0
B
63.0 ± 25.0
AB
66.7 ± 15.0
AB
80.0 ± 10.0
A
96.7 ± 5.8
A
7 days
30.0 ± 30.0 B
83.0 ±15.0 A
73.0 ±21.0 A
66.7 ± 21 .0 A
93.0 ± 5.8 A
93.0 ± 5.8 A
 Product Degradation

 TABLE 4.    Degradation of D-Limonene under Field Conditions.
             Samples were taken from plots treated with one of four concentrations of
             d-limonene. Analysis was performed by GC/MS.
Concentration
1.5%
3.0%
5.0%
10.0%
Quantity of D-limonene
Day 0
0.053

0.088
0.036
Day 1
1.240
0.309
0.237
0.983
Day 2
0.288
0.125
0.265
5.000
mg/kg soil)
Day 4
0.010
0.020
0.010
0.070
Day 6
0.002
0.008
0.007
0.002
                                       53

-------
 COST/BENEFIT ANALYSIS

      The above results show that d-limonene is as effective as a leading toxicant,
 Orthene®, in controlling fire ants. Currently available information indicates that all
 Dead Ant ingredients are non-toxic to humans and other vertebrates and are not
 harmful to the environment.  In addition to being safe and effective, Environmental
 Pesticides Group expects to offer Dead Ant at a price competitive with Orthene® and
 other leading fire  ant control products.  The approximate retail cost of Orthene® is
 $0.25 to treat a single mound.  Currently, Dead  Ant can be produced at about $0.12
 per mound treatment (1 to 1.5  pints per 6- to 8-inch mound), and the anticipated retail
 price is $0.22 to $0.25 per mound treatment.


 CONCLUSIONS

 POLLUTION PREVENTION ASSESSMENT

 Incentives

      The product is almost ready to be introduced into the marketplace. The major
 hurdle remaining is EPA registration of the product, which requires extensive tests on
 animal toxicity and environmental fate.

      The targeted markets will include farmers, nursery and sod growers, park
 maintenance personnel, pest control operators,  gardeners, and home owners -- all
those concerned with the control of fire ants.


 Barriers

      The only shortcoming is  an occasional, temporary yellowing of surrounding
grass following treatment of mounds.  Environmental Pesticides Group is testing
different concentrations of Dead Ant under various  conditions (e.g., grass type and
temperature) in order to find the combination of conditions that will eliminate  grass
discoloration and still provide effective fire ant control.
                                      54

-------
                           POLLUTION PREVENTION
                             IN CADMIUM PLATING

                                      by

                              Mandar Sunthankar
                              lonEdge Corporation
                             Fort Collins, CO 80526
ABSTRACT
      Cadmium is electroplated on many industrial components because of its
desirable lubricity  and corrosion resistance properties. However, the use of cyanide
baths in electroplating and toxic waste disposal related to cadmium are of significant
environmental concern.  In recent years, 50/50 zinc-cadmium alloy coating has shown
promise as an alternative to cadmium. This alloy uses 50 percent  less cadmium in
plating and exhibits corrosion properties superior to those of cadmium.

      To minimize environmental and occupational hazards related to cadmium
electroplating, a novel dry plating technique has been developed by the lonEdge
Corporation. This plating eliminates liquid chemicals and prevents solid waste using in
situ reclaim. The  dry plating process  is suitable for plating the 50/50 zinc-cadmium
alloy. In order to  explore the commercial potential of this dry plated alloy, its lubricity
was studied and compared with that of the electroplated cadmium.

      In the research conducted, the 50/50 zinc-cadmium alloy was dry plated on
mild steel washers.  Sample washers  were also electroplated for comparison. These
washers were used as substrates for  the lubricity study.  The lubricity was measured
in terms of coefficient of friction using pin-on-disk method. The chemical composition
of the dry plated alloy was determined using Energy Dispersive Spectrometric (EDS)
x-ray analysis.   The data indicated that the coefficient of friction of  dry plated 50/50
zinc-cadmium alloy is 0.133 compared to 0.127 for electroplated cadmium. The
statistical t-test  of  significance predicts that this difference is not significant. As a
result, lonEdge concluded that the lubricity of this alloy is competitive with that of
cadmium and superior to that of known values of zinc.  This lubricity study indicates
promising commercial potential for the dry plated 50/50 zinc-cadmium alloy.


INTRODUCTION

PROJECT DESCRIPTION

      Electroplated cadmium is preferred in many large-scale fastener applications
because of its superior lubricity and corrosion  resistance in marine environments
compared to that  of other coatings.1  In cadmium  electroplating operations, exposure
to toxic cadmium  dust is a safety issue, and disposal of  cadmium  sludge is an
environmental  issue.2  Consequently,  there is substantial interest in reducing the use of
cadmium and related waste. However, two decades of extensive effort in this area
has yet to result in an effective  economical solution.
                                      55

-------
      The dry plating process developed at the lonEdge Corporation was proposed
as exhibiting promising pollution prevention potential. Dry plating eliminates use of
liquid chemicals and prevents solid waste using in situ reclaim. In recent years, a
50/50 zinc-cadmjum alloy has been qualified by the automobile industry for some
fastener applications.3  Environmentally safer zinc plating alone does not exhibit the
frictional property required to replace cadmium. The frictional coefficient of zinc (0.16)
is relatively higher than that of cadmium (0.123). The addition of cadmium is likely to
enhance the lubricity of zinc and bring it into an acceptable range for fasteners.  The
proposed effort uses a non-liquid chemical process for plating this lower-friction zinc-
cadmium alloy and  studies its frictional property with respect to cadmium.

      A study of the lubricity of a dry plated 50/50 zinc-cadmium alloy is desirable for
its potential  application in a variety of industries! The following objectives were
targeted for this research:

      Measure the  lubricity of dry plated 50/50 zinc-cadmium alloy in terms  of
      coefficient of friction.

•     Compare the coefficient of friction of dry plated 50/50 zinc-cadmium with that of
      electroplated cadmium.

Outline of Process/Product

      Cadmium is a known toxic material.  Occupational exposure to cadmium dust
and vapors  is a safety issue, and disposal of cadmium is an environmental issue.2
Most cadmium electroplating is carried out in cyanide and other toxic chemical baths.
Alternative coatings of zinc or aluminum  do not exhibit certain unique combinations of
properties of cadmium.4'5 Consequently, despite escalating costs of  waste
management from stricter environmental regulations, cadmium electroplating continues
to be widely used in defense, aerospace, and automobile industries.

      There is substantial interest in reducing cadmium in plating.2  A new coating
recently qualified by the Ford Motor Company is a 50/50 zinc-cadmium alloy3, (i.e., 50
percent by weight zinc in cadmium (Ford Specification S54-M). This alloy contains 50
percent less cadmium in deposited coatings.  The 50/50 zinc-cadmium alloy shows
suitable combination of properties of both zinc and cadmium, and is  superior to
cadmium in  corrosion properties.3 The outdoor exposure studies of competing
coatings in various environments indicated that zinc-cadmium alloys outperform other
coatings in every environment tested.6 The lubricity properties of zinc-cadmium alloys
were not reported.

      Various compositions of the zinc-cadmium alloy coatings are currently applied
using the mechanical barrel plating process.3'6  This plating is a liquid chemical
operation free of cyanide and chelates.7  However, significantly larger amounts of
cadmium and other toxic chemicals are discharged following each batch of plating
compared to electroplating. This results in a cost intensive operation.8

      A plating method that minimizes the use of cadmium as well as the use of toxic
liquids is desirable.  The dry plating addresses this need  with three unique features:
                                       56

-------
      The plating rate is competitive with electroplating.
      Solid waste is minimized using in situ reclaim and recycle.
      Furthermore, the zinc-cadmium alloy would reduce the amount of cadmium
released into the environment by discarded components.

Unique Features of the Proprietary Dry Plating Process .

      In the dry plating of zinc-cadmium alloy, simultaneous plating of zinc and
cadmium species is conducted under a neutral gas glow discharge condition. The
glow discharge offers the following advantages:

•     The collision of ionized neutral gas molecules with zinc and cadmium particles
      results in atomic mixing of the species before plating.

•     The bombardment of the ionized species  on the substrate enhances adhesion
      of the growing layer to the substrate.

•     Energetic particle bombardment of the surface densifies the alloy layer.

      Details of this process cannot be disctosed due to the proprietary nature of this
method.  A schematic of the setup is shown in Figure 1.  The scattered, non-
directional nature of the plating molecules produces excellent plating uniformity without
rotation of the parts.  In dry plating, surface cleaning before plating is performed in the
same process cycle using neutral gas ion bombardment.  There are no by-products or
waste generated in this "dry" cleaning. In addition, build-up of coating at edges and
corners is avoided. Plating rates competitive with electroplating have been
accomplished  in the dry plating process.  Typical cycle time for a 0.5 mil-thick plating
is less than 10 minutes.
Figure 1.    Dry Alloy Plating Apparatus
                                  ELECTRIC BIAS

ELECTRODE —
ZINC SOURCE —
ZINC MOLECULES —


ELECTRODE —



t
„,»», ,}„,„,»
m ,<~^ m
-4*. /* .j^S^v \ «*f-
i !». / GLOW \ „! •
!^ \ DISCHARGE j ^|
| — r-j
4^^^NS^^^NN^^^^^^^l
^^ \
IN ^ \


	 CHAMBER
	 CADMIUM SOURCE
UAUMIUM MULtCULES





                                             WASHER
                                       57

-------
Summary of Prior Studies Conducted

      Preliminary studies, such as salt fog corrosion test (ASTM B117), adhesion test
(ASTM B571), and solderability test using pull method, conducted on dry plated
cadmium coatings have been promising. In addition, other studies have been
completed under the Small Business Innovative Research (SBIR) program.  In a study
sponsored by the National Science Foundation (NSF), it was demonstrated that the
dry plating rate can be varied over nearly two orders of magnitude, from 0.01 mil per
minute to 0.7 mil per minute.  In a Department of Defense-sponsored study, it has
been demonstrated that the plating uniformity on complex shapes is competitive to
electroplated counterparts.  This uniformity also surpassed required military standards
according to specification MIL-C-8837B.

Technology Advantages

      The extraneous metal deposits in the dry plating chamber will be continuously
or periodically recovered or transferred to the metal source itself using  a proprietary
technique.

      Occupational exposure of an operator to hazardous dust and fumes is an issue
of concern in a plating operation. The dry plating process eliminates this exposure.  A
study was also conducted according to ASTM D4185 using Atomic Absorption
Spectrometry (AAS) to determine the level of cadmium or zinc dust emission from the
dry plating equipment.  This study showed that the level  of cadmium and zinc
emission was below the AAS detection limit (i.e., below 50 parts per billion:  10"9 grams
per cubic meter). This  is comparable to the levels of these metals measured in non-
polluted air. In addition, dry plating eliminates waste and sludge disposal.

APPLICATION

Process/Product Replaced

      There are 1,166 facilities in the U.S.  predominantly electroplating  cadmium.2
The large industry of cadmium platers and users is well established with decades of
proven procedures and specifications.  This industry requires new yet compatible and
economical methods of waste reduction.  Electroplated cadmium is preferred when
users need corrosion protection in a marine environment, or when coating lubricity is
of prime concern for components like steel fasteners.1  For example, Ford Motor
Company uses 30 million cadmium-plated  nuts annually in addition to millions of other
components; 50,000 to 60,000 pounds of components are cadmium-plated  daily in
other plating shops.3 Electroplated cadmium typically used is under 0.5 mil (13ji/m)
thickness.

Current Cadmium Plating Methods

      Three distinct methods  are used in plating cadmium or zinc under 1 mil thick:
electroplating, mechanical plating, and vacuum cadmium. Cyanide bath electroplating
has been the overwhelming choice for most applications  because of good quality and
high throughput at reasonable cost.1 Mechanical plating  is limited to barrel  plating of
small components, and is more competitive for thicker  coatings (>0.5 mil).8  Vacuum
cadmium is expensive and is preferred only when hydrogen embrittlement has to be


                                      58

-------
eliminated in some defense and aerospace applications.9  Barrel plating is used in
plating small parts, such as fasteners in bulk (2,000 to 10,000) in a rotating barrel.
Rack plating is used for individually plating larger parts on racks.

Occupational Hazards of Cadmium

      Occupational exposure to cadmium primarily occurs through inhalation of dust
or fumes.2 According to EPA's 1984 guidelines, cadmium is classified as a group B1,
probable human carcinogen.2 Current OSHA permissible limits for 8  hour cadmium
fume exposure is 0.1  milligrams per cubic meter and that for cadmium dust exposure
is 0.2 milligrams per cubic meter (29 CFR 1910.1000).  The current EPA standard for
limiting the amount of direct cadmium discharge in wastewater on a monthly average
is not to exceed 0.26 milligrams per liter (40 CFR 433).

Other Environmentally Safer Coatings

      Zinc exhibits many properties similar to cadmium.  Zinc plating is relatively
inexpensive and has  been used to replace cadmium where appropriate.10 However,
as can be seen from  Table 1, cadmium has better lubricity and  torque coefficient and
is more corrosion resistant in marine environments.
TABLE 1.    Comparison of Cadmium and Zinc Plaging Characteristics
Characteristic
Typical thickness
BtJtter corrosion
resistance in:1
Corrosion product1
Coefficient of friction1'11
Lubricity10
Torque coefficient10
Cadmium Plating
0.2-0.5 mil
Marine environment
Thinner, ductile, adherent
Lower than steel (0.12)
Excellent
0.15-0.25
Zinc Plating
>1.0 mil
Industrial environment
Thicker, harder, loose
Higher than steel (0.16)
Fair
0.21-0.33
      Low torque coefficient is preferred especially in automated fastener assembly
operations.  In this regard, the lubricity of dry plated 50/50 zinc-cadmium alloy is likely
to be superior to that of zinc and closer to that of cadmium.

      A study conducted by the U.S. Army indicated that the combination of
corrosion resistance and lubricity of cadmium was superior to 14 other competing
materials including aluminum.12 According to two other studies conducted by the U.S.
Air Force, aluminum is useful at higher temperatures and is superior in abrasion
resistance.4-5  In such applications, aluminum could replace cadmium.  However, for
equal thickness, cadmium performed better in salt fog corrosion  tests.4  Close
tolerance in fastener applications require thinner, ductile coatings of cadmium.
Electrically non-conductive corrosion products of aluminum are unacceptable in the
                                      59

-------
 electrical connector market.  The complexity in aluminum plating operation and high
 capital cost made it unlikely that it will ever become common in plating shops.13

 Wastes Prevented

      Conventional electrochemical plating processes use several hazardous liquid
 chemicals in large quantities.  These include alkaline cyanide baths, and various
 chlorides, sulfates, phosphates, and fluoborates of heavy metals. On the contrary, dry
 plating does not use any liquids. In summary the dry plating method offers the
 following benefits in that it is a simple pjating operation;  recycles pure solid metal in
 situ; eliminates hazardous liquids used in plating; eliminates  subsequent wastewater
 treatment and sludge; reduces costs related to chemical equipment; and reduces
 capital costs and operating space.  Since no liquids are required in plating, the
 subsequent costs and liabilities of waste treatment,  disposal, and effluent discharge
 are eliminated.

 Cadmium Reclamation from Discarded Components

      In cadmium  plating shops, plated components that are rejected for poor quality
 are stripped in toxic liquids before replating.  Instead, dry plating equipment can be
 used to reclaim this cadmium as well as that from used discarded components before
 incineration. Contaminated cadmium reclaimed this way can be recycled by the
 cadmium producers.

 Cross Segment Uses

      The initial  niche of dry plating technology is likely to be in steel fasteners and
 electrical connectors used on a large scale in the Department of Defense and the
 automobile industry.  Current trends in these segments indicate increasing use of
 alternatives to  cadmium electroplating.  Another potential segment for this technology
 is likely to be the high strength steel components used in the defense and aerospace
 industry, because dry plating will eliminate hydrogen embrittlement and the process
 will be very competitive.  Available information suggests that  among high strength
 steel users, the U.S. Navy and Air Force are leading the  effort in seeking alternatives
 to cadmium electroplating.

      Dry plating will also plate zinc without using liquid chemicals. As  an alternative
 to zinc electroplating (galvanizing), dry plating has potential in strip-steel  plating and
 wjre coating. Considering the high  rates of dry plating, galvanizing steel sheets and
 wires may be done economically.  Dry  plating also can galvanize sheet metal and wire
 products after  metal finishing.


 PROCEDURE

 DEMONSTRATION  PROCEDURES

      lonEdge used two independent experimental  procedures to complete this
project:  (1) the alloy plating plus composition analysis; and (2) the lubricity
measurements. The 50/50 zinc-cadmium alloy deposition process needed some
improvements  so that the same composition could be plated repeatably.  In this


                                      60

-------
development phase, smaller-sized mild steel substrates were plated so that these
could be placed in composition analysis equipment. The composition analysis was
performed using Energy Dispersive Spectrometric (EDS) x-ray analysis. Later, larger,
medium carbon mild steel washers were plated for lubricity study.  Washers are
frequently used for frictional (lubricity) tests and are commercially available in various
sizes. Lubricity was measured on a pin-on-disk frictional test set-up. The pin and the
disk (washer) are made of the materials to be tested against each other. In this case,
the interest was in the performance of the alloy coating against mild steel.
Consequently, the washer was alloy plated  and the pin was made of mild steel.

The Dry Plating Apparatus

      The basic principle of the alloy plating process is as follows: A desired
composition of the alloy is obtained by proportioning the quantity of zinc and
cadmium species injected in the chamber.  This quantity is proportional to the amount
of source material to be plated.  To minimize complexity in the process, the source
temperature is maintained nearly constant to allow for  better control over the process
and produces acceptable repeatability. The desired composition is obtained by
keeping the injected quantity of zinc species constant  while varying the injected
quantity of cadmium species.  The process parameters are manipulated to vary the
amount of metal species in the injected stream. The plated samples are EDS
analyzed. This exercise helped to understand the performance of the plating process
and its sensitivity  to the process conditions. lonEdge  cannot disclose further details of
this process due to its proprietary nature.

      The glow discharge apparatus used  for dry plating is made by Technics (Model
Hummer V). The minimum pressure obtained by a 1.5 horsepower mechanical pump
in this apparatus is about 20 milliTorr.  Zinc and cadmium sources were 99.5 percent
pure and were purchased from a commercial vendor.

Sample Preparation for Alloy Plating Tests

      Medium carbon steel substrates were used to simulate typical practical
conditions where  cadmium coatings are used.  For the initial composition experiments,
the substrates were made from available commercial grade, cold-rolled mild steel,
0.195 centimeters thick.  These were cut into approximately 1.15 centimeter squares
and ultrasonically degreased in acetone and methyl alcohol. All substrates were
polished on 600 grit SiC sandpaper and subsequently polished to mirror finish using
0.3 micron alumina slurry. This same degreasing  and  polishing procedure was used
for as-purchased  mild steel  commercial washers.

EDS X-ray Analysis

      This quantitative chemical analysis was performed in the Scanning Electron
Microscope (SEM) by measuring the energy and intensity distribution of the x-ray
signal generated by a focused electron beam. A Phillips 505 SEM with secondary
backscatter and windowless x-ray detector,  and KeVex Super 8000 EDS spectrometer
with associated Quantex software was used. During measurements, the incident
electron beam was set at 15 degrees to the normal to the substrate, and the x-ray
detector sensed characteristic x-rays emitting from the surface. To minimize
measurement  error due to microscopic variations  in composition of inhomogeneous


                                      61

-------
zinc-cadmium alloy, a 0.2 x 0.2 millimeter area was scanned and quantitative
measurements were averaged.

The Lubricity Test Set-up

       The pin-on-disk set-up consists of a balanced beam (arm) at the end of which a
mild steel pin is mounted in a  bushing.  This pin rests on a washer mounted on a
cylindrical aluminum block. The washer is plated with the desired coating. There is
zero weight on the pin before  the start of the test.  A known weight is placed on the
pin as a normal load.  The washer is turned  by  a motor (Danfoss, Model Varispeed
2000) at a desired speed (typically 100 to 1,000 revolutions per minute). The frictional
force generated at the contact between  the pin  and the coating is displayed on a
bridge amplifier/meter  (Ellis Associates).  The test is run for a few hundred to 10,000
rotations (cycles) depending on the strength of  the materials being tested.

Sample Preparation for Lubricity Study

       The mild steel pin  was made in-house from a 0.625 centimeter diameter stock
material. The tip  of this pin was machined to a  hemispherical shape using a cutting
tool, and the tip was polished  to mirror finish on a buffing wheel using  0.3 micron
alunnina slurry.  The mild  steel  washers selected for the lubricity test were 3.75
centimeters in diameter and 2  millimeters thick.  These washers were degreased and
ground to an acceptable flatness and polished on a 600 grit SiC sandpaper. These
were plated with the desired material and mounted on a 5 centimeter diameter and 5
centimeter long aluminum cylindrical block during tests.  Five polished  washers were
sent to Denver Metal Finishing Company, Denver, Colorado, for conventional bright
cadmium electroplating.  The lubricity of these was compared later with that of the dry
plated washers.

Pin-on-disk Apparatus  Calibration

      According  to the standard calibration procedure, the weight on the pin due to
the balanced beam before calibration was zero  (i.e., the arm movement was nearly
free of weight and friction around the pivot. The tester was calibrated using standard
weights in grams. During dynamic frictional studies, the frictional force at the interface
of pin and washer created a drag (load) that resulted in minute tension and bending
of the arm.  This bending induced by the load, called frictional force, is proportional to
the friction coefficient.  A  sensitive tensile test meter indicated the extent of bending in
terms of divisions on an arbitrary scale.  This bending  can also be induced using
standard weights in static mode. The scale was correlated to standard weights that
indicate frictional force. Accordingly, standard weights were applied to the arm and a
calibration curve (Figure 2) was developed for the apparatus.  Using this curve, the
frictional force was extrapolated from deflection  of the  meter during dynamic tests.
The formula for Coefficient of Friction is:

            Coefficient of Friction  = Frictional Force
                                Normal Load


where Normal Load is placed on the pin during  the dynamic test.
                                       62

-------
Figure 2.    Calibration Curve, Deflection versus Frictional Force
oc
o
z
tu
o
c
£
          O

          O
          £
          u.
               20 --
               15 --
               10 --
     5 --
                 05         10        15       20

                 TENSILE TEST METER DEFLECTION IN DIVISIONS
       Frictional tests were run without using a lubricant to simulate typical situations
for cadmium plated fasteners.  In the case of hard coatings, such as diamond,
nitrides, and carbides, weights in excess of 200 grams typically are used as normal
loads on the pin,  and washer rotational speeds are in hundreds of revolutions per
minute. For soft surfaces like cadmium and mild steel, this procedure was too severe.
The metal was immediately seized, gouged, and penetrated.

       Consequently, the following acceptable procedure was developed over a series
of experiments. The frictional values were compared to those published. A 20 gram
weight was placed on the pin as normal load.  The washer was turned by a motor at
50 revolutions per minute. The frictional force generated at the contact between the
pin and the washer surface was displayed on the tensile test meter.  This test was run
for 100 rotations,  or 2 minutes,  on the washer.  Longer time or larger loads were not
possible for most samples.  In the case of fastener applications, only a few turns are
required for tightening.  Consequently, 100 turns should provide adequate simulated
data for multiple fastening of the parts.
                                       63

-------
 Cost of Demonstration

       The research portion of this project was conducted in the Materials Laboratory
 at Colorado State University (CSU) according to a prior agreement. The total cost of
 the project, including labor, overhead, materials, and equipment was $27,025  with
 $25,000 provided by EPA through the Pollution  Prevention By and  For Small Business
 Grant, and the remaining $2,025 provided by lonEdge.  The direct  labor included the
 Principal Investigator and one Mechanical Engineering undergraduate as a part-time
 technical assistant.  All necessary equipment was located at CSU and was made
 available for research, thus there were no costs to use these items except for  repairs
 For accurate EDS data and analysis, the zinc-cadmium alloy-plated test samples were
 sent to an outside analytical service under subcontract.

       The time required to develop a procedure to adjust the ratio of zinc and
 cadmium in the alloy was approximately six weeks longer than the  projected two
 months due to high sensitivity of cadmium species to the process parameter
 variations.  This problem was corrected by regulating the supply of cadmium species
 entering the chamber.  The lubricity study was completed in the scheduled two
 months.  During the last month of experiments, the anode power supply in the glow
 discharge unit broke down due to auto-transformer overload. This required repairs
 arid caused a three week delay. Total time to complete this project was approximately
 six months.


 REESULTS AND DISCUSSION

 PERFORMANCE RESULTS

 50/50 Zinc Allov Plating

      These plating experiments were conducted on smaller size, 1.15 square
 centimeter mild steel substrates to avoid sectioning and excessive handling of
 washers. The smaller substrates could be easily placed inside the SEM chamber
 The EDS x-ray composition analysis indicated that zinc content was high initially (zinc
 = 62 percent by weight) as indicated by Figure 3, sample #7.  As the cadmium flow
 was increased by changing process parameters, a larger than expected shift in favor
 of cadmium content in the alloy (cadmium = 77  percent  by weight)  was noticed for
 sample #E2-4.  This EDS data is presented in Figure 4.  The process parameters
were adjusted  in successive experiments and composition was analyzed  The  quantity
of cadmium in the alloy was lowered to 49 percent as shown in Figure 5 of sample
   ~\ i.
                                     64

-------
Figure 3,    EDS Analysis of Dry Plated 62/38 Zinc-Cadmium Alloy
                                                                        9
                                                                       X.
                                       65

-------
Figure 4.    EDS Analysis of 77 Percent Cadmium, Dry Plated Zinc-Cadmium Alloy
                                                                           9
                                                                           Jf.
                    o;
                    UJ
II-
•<
                                                                            .0
                           in  to
                           o  o
                           o  o
                           ^  0)
                            t   •
                           o  o
                                                               o
                                                               o
                                                                      Q.

                                                                      0
 9
 Q
 c
 •

 9
<*.
                                                                                   i
                                                                                  CSJ
                                                                                  W
                                                                                   CL


                                                                                   03
                                      66

-------
Figure 5.     Dry Plated Zinc-Cadmium Alloy with 50 Percent (± 2 Percent) Zinc or

             Cadmium
               ?i
               38
                           ^ cn
                           CD cn
                           CD cn
                           cn CD

                           o d
                           cn  cn
                           in  o>

                           ^  o
                  9
                  4*


                  D
                     Z
                     Ui
                     U
                     cc
                     UJ
                     0.
                     UJ
                     z
09
      cn  o
       •   •
      <*  in
      o  cn
                            CD  CM o
                            ^  CD o

                            ~  co d
                            in  ^ o
                       CO
                            <  <
       c  TJ
       N  u
-H

 O

 CM
<


O
                                                                              GO
                                                          i

                                                          CM

                                                          W
                                                          C.


                                                          0}

                                                          CO
                                                                              cn
                                                                              CM
                                    cn
                      O
                      cn
               S
               ca
                                                         CO
                                          CJ
                                         67

-------
       The composition of alloy dry plated on sample #E2-11 being within acceptable
 limit (± 2 percent) of the 50/50, it was concluded that the desired zinc-cadmium allov
 was dry plated.                                                                J

 The Lubricity Study

       Three washers were also dry plated with the 50/50 zinc-cadmium alloy under
 the same process conditions.  Comparative lubricity studies were conducted on plain
 steel washers, cadmium electroplated washers, and 50/50 zinc-cadmium dry plated
 washers. Lubricity was measured in terms of coefficient of  friction as described
 earlier.  The average of frictional measurements for electroplated cadmium (0 127) was
 quite close to that reported in the literature (0.123).11  This indicated that the error in
 the measurement or procedure was relatively small.  This low value of cadmium
 frictional coefficient is preferred in the fastener industry.  The other data presented in
 Table 2 indicate that the frictional coefficient of the dry plated  50/50 zinc-cadmium
 alloy is slightly higher than that of electroplated cadmium.  However, this difference is
 statistically not significant (as explained later) and is  likely to be in the range of
 acceptable values for steel fasteners.  This result is presented in Figure 6 using a bar
 graph.
TABLE 2.
Lubricity of Mild Steel Washers, Cadmium Plated Washers and Allov
Plated Washers                                        '
Washer
Type
No coat
No coat
=s=^=^=
EPCad
EPCad
EPCad
Zn-Cad
Zn-Cad
Zn-Cad
^— —
Sample #
PS-1
PS-2
DM-2
DM-3
DM-4
50Cd-1
50Cd-2
50Cd-3
Deflection
(Divisions)
3.4
1.5
0.8
1.0
2.0
1.0
2.0
1.0
=
Frictional
Force
6.8
3.0
1.6
2.0
4.0
2.0
4.0
2.0
=====±
Frictional
Coefficient
0.34
0.15
0.08
0.10
0.20
0.10
0.20
0.10
Average
Coefficient

0.245





0.133
                                      68

-------
Figure 6     Bar Graph Showing Lubricity of Mild Steel Washers, Cadmium Plated
            Washers, and Alloy Plated Washers
                                                  STEEL vs. 50/50 Zn-Cd
                                                  STEEL vs. EP CAD
                                                  STEEL vs. STEEL
                     0.06         0.1         0.15

                         COEFFICIENT OF FRICTION
                                                      0.2
                                                                 0.25
 PRODUCT QUALITY VARIANCE

       The appearance of the dry plated substrate surface was dull metallic gray
 compared to the bright metallic of the electroplated counterparts.  Two factors were
 likely to influence this appearance.  In the electroplating process, brighteners are
 added to the bath to give the part surface added shine. Second, the morphology of
 the dry plated surface is granular and consequently non-specular, as the SEM
 micrograph indicates.  A small degree of preferential crystal growth appears to have
 occurred during plating.  This is not uncommon in low pressure processes and can be
 corrected with process modifications. However, the grainy surface of this ductile metal
 is not likely to influence the frictional properties because the surface is smeared easily
 as soon as the test begins and the  structure is altered. This is also true with the
 electroplated cadmium surface.
                                       69

-------
        Significant variation in frictional values of plain polished steel washers was
 attributed to the variation in hardness from washer to washer. When the material
 hardness was relatively lower, the wear track was abraded and galled, and the friction
 was greater.  The debris could be observed on the tip of the pin under the
 microscope.  When the material was harder, little debris was observed, and the wear
 track was less damaged.  On the other hand, the wear track on the electroplated
 cadmium, as well as on the dry plated alloy washer, was also damaged soon after the
 test was resumed.  Similar debris could be observed  on  the pin tip.  However friction
 was low and  its variations  were relatively small due to the soft and ductile nature of
 these materials.  In other words,  lubricity was  much better,  and seizing was prevented
 This property is valuable for fasteners that need consistent and lower torque values in
 automated assembly operations.

       The statistical information in Table 3 compares  the coefficient of friction of drv
 plated  and electroplated samples.


 TABLE 3.    Statistical Information Regarding Coefficient of Friction of Dry Plated and
             Electroplated Samples

Mean
Standard Deviation
Variance
Sample Size
Dry Plated 50/50 Zrt-Cd
x, = 0.133
s1 = 0.058
s,2 = 0.003
n1 = 3
Electroplated Cadmium
x, = 0.127
S9 = 0.064
S92 = 0.004
a, = 3
       The variation is frictional values of dry plated samples is slightly smaller than the
 electroplated counterpart.  In order to estimate if the difference in the mean would be
 statistically significant for the large population based on the small sample size the t-
 test of significance was conducted. This analysis indicates that, at 99 percent
 confidence level, there is no statistically significant difference between the lubricity of
 dry plated 50/50 zinc-cadmium alloy and electroplated cadmium. lonEdge concludes
 that the dry plated 50/50 zinc-cadmium alloy has the potential to replace electroplated
 cadmium.                                                                 ^

 CONDITIONS THAT IMPACT PERFORMANCE

      The lubricity, or frictional coefficient, of a material is a surface phenomena
 Consequently, surface contamination could influence the lubricity property  Similarly
 changes in physical properties of materials, such as hardness and crystal structure
 could result in changes  in lubricity. Soft materials like cadmium, zinc, and graphite are
 relatively less sensitive to these factors except for severe surface contamination  For
 example, zinc oxide is a corrosion product of zinc that accumulates on the  metal's
 surface after a period of exposure to the environment.  This oxide has poor lubricity
which is unacceptable in the fastener industry.  Consequently, zinc coating  is avoided
for critical fastener applications.
                                       70

-------
      In the case of dry plating, pure metals are plated, and surface contamination is
avoided due to vacuum environment. Also, absence of liquids in dry plating will
prevent surface or bulk contamination that can occur in electroplating.  The other
factor affecting lubricity is the zinc to cadmium ratio in the plated alloy.  If the
composition is held  within  ± 2 percent of the desired value, the lubricity change is not
likely to be noticeable.

COST/BENEFIT ANALYSIS

      The following data present waste generated and costs incurred in the
electroplating operation that is targeted for elimination  using the dry plating process.

Costs Due to Environmental Regulations

      Cadmium electroplating operations use large amounts of chemicals and water.
A typical job-shop discharges  between 60,000 to 150,000 gallons of effluents per day,
and may use over 20,000  pounds of cadmium per year.9-14715  In order to comply with
EPA regulations, the installation cost for water pollution control system and associated
hardware exceeds $650,000.9>16 Furthermore, the daily cost of wastewater treatment in
a typical medium size facility is approximately $1,200 per day including the cost of 15
to 20 tons of toxic sludge disposal per week.9 As lower discharge limits are being set
on heavy metals and as the number of sludge disposal sites is diminishing, the cost of
waste treatment is escalating.17

Anticipated Waste Reduction Cost Benefits

       Waste treatment costs in cadmium electroplating are outlined  in Table 4. These
are typical estimates of a few job  shops studied.4  These costs will be eliminated by
the dry plating process.  Cadmium waste generated, if any, will be in solid metal form
that can be reclaimed and recycled.

TABLE 4.    Electroplating Waste Treatment  Costs4
Cost Factor (hourly)
Recurring
Labor
Chemicals
Disposal
Utilities
Miscellaneous
Capital
Total Cost
Cost of Waste Treatment Facility
$86
38% - maintenance, analysis, records
1 6% - cyanide treat, Cd precipitation
17% - haz. wastes, Cd sludge, filters
26% - water, electric
3% - OSHA compliance, permits,
insurance
$42 (7 year amortized; 2,000
hours/year)
$1 28/hour or approx.
$1 ,000/day/facility/shift
                                        71

-------
 Anticipated Labor Cost Savings in Dry Plating

       The dry plating operation will use the same process flow as that of vacuum
 cadmium plating. The significant difference, however, will be in labor cost savings as
 shown in Table 5.  The dry plating cycle being competitive with electroplating in
 deposition rate and time, labor costs in the actual plating cycle were assumed to be
 the same As the data suggest, dry plating is likely to save nearly 30 percent in labor
 costs and will be the least labor intensive process.


 TABLE 5.    Process Flow Chart (Labor in Minutes)4(except dry plating)
^===^ 	
Vacuum Cadmium
Vapor degrease (14)
Grit blast (32)
Evaporation coat (40)
Inspect (6)
Chromate conversion
Coat (4)
Water rinse, dry (6)
Inspect (4)
Total labor (106 minutes)
=========================^^
Electroplated Cadmium
Vapor degrease (1 4)
Grit blast (16), alkaline
clean, water rinses, and
pickle (25)
Electroplate (10), water
rinse, dry (7)
Inspect (6), alkaline clean
and rinses (10), dip in
cadmium plate and rinse
(5)
Chromate conversion
Coat (4)
Water rinse, dry (6)
Inspect (4)
Total labor (107 minutes)
.
Dry Plating
Vapor degrease (14)
Grit blast (32)
Dry plate (10)
Inspect (6)
Chromate conversion
Coat (4)
Water rinse, dry (6)
Inspect (4)
Total labor (76 minutes)
 Cost Benefit Summary

      The dry plating process could save $1,000 per day per shift in waste treatment
 costs and reduce labor costs by 30 percent compared to conventional electroplating


 CONCLUSIONS

 POLLUTION PREVENTION ASSESSMENT

 Incentives for Dry Plating of 50/50 Zinc-Cadmium Alloy

      Based on the coefficient of friction data and microscopic observations l
concluded that the lubricity of dry plated 50/50 zinc-cadmium alloy \s campe
                                      72

-------
the electroplated cadmium.  Previously reported data Indicate the superior corrosion
performance of zinc-cadmium alloys in most environments.6  Such a study for dry
plated zinc-cadmium alloys needs to be  conducted. However, since corrosion
resistance is a basic material property, it is unlikely that it will depend on the type of
process used for plating, unless the coating is highly porous and/or contaminated.
The dry plating process plates metals  of purity superior to electroplating. These facts
lead loriEdge to believe that the dry plating of 50/50 zinc-cadmium is a promising
alternative to cadmium electroplating, as it has the potential to reduce waste in the
fastener and connector plating industry.  50/50 zinc-cadmium alloy will reduce waste
during disposal of discarded components.  Also, the technology has the potential for
setting a new standard for occupational  safety.

      As a  result of the features and advantages of the dry plating process, several
federal agencies, including the EPA (through this grant program and the Small
Business Innovative Research program), have contributed to this technology. The dry
plating project is now in its third year.  During this time, a utility patent was filed with
the U.S. Patent Office, and a prototype batch plating apparatus is under construction
for cadmium, zinc, and zinc-cadmium  alloys.  Research facilities, expertise,  and
analytical support have been provided by the Materials Laboratory of the Colorado
State University under a collaborative agreement.

      At this time, three potential plating industry segments are being considered for
the dry plating technology transfer. These are: discrete components, such as
fasteners and connectors plated for a broad segment of users; strip-steel plating for
the steel industry; and steel wire coating for the construction industry. Although the
current effort is concentrated on discrete components and batch plating apparatus,
this technology has considerable potential in continuous galvanizing of strip-steel and
steel wire.  This is feasible because of the high plating rates achievable in dry plating.
Two private plating companies  have made commitments for collaboration in the
current phases of technology development as well as for the next phases of
commercialization.  The apparatus deyelopment  is  being tailored to the needs of these
companies, each company being in different applications areas.  Similarly,  new
customers will be sought for collaboration in other  market segments, and new
apparatus will be developed according to their needs.  The commercial use of this
process  will begin in mid-1994.


Potential Barriers to the Dry Plating Technology

       Dry plating needs to  be tested  in pilot-line environments to resolve operational
difficulties that new technologies encounter. Some products may prove to be more
profitable than others.  Like electroplating, the same equipment may not be suitable
for barrel plating and rack plating.  For  example, continuous plating of wire or sheets
will require a different configuration of the dry plating apparatus.

       One  solution to these problems is to set up dry plating equipment within an
existing electroplating line, with the plating tank paralleled by a dry plating  apparatus.
Some products can then be transferred to  dry plating over a period of time.
Economics and customer satisfaction can be resolved during this period.
                                        73

-------
      The many challenges described here have been resolved for electroplating over
 many decades.  Dry plating is not likely to solve such difficulties in short term, as well
 However, as compared to electroplating, dry plating is a relatively lower risk, high pay-
 off, cutting edge technology that can benefit those who are seeking alternatives to
 conventional technologies.

 REFERENCES

 1.    Metals Handbook, American Society for Metals, 9th Edition, 1982.
 2.    Federal Register, Vol. 55, No. 25, February 6, 1990, OSHA Proposal for
      Cadmium.
 3.    Meyer, W.T., SAE Technical Paper, International Congress & Exposition Detroit
      Ml, February 1986.
 4.    Holmes, V.L et al, AFESC Report, August 1989, DTIC Publication #AD-A215
      633.
 5.    Carpenter, C.J., AFESC Report, March 1988, DTIC Publication #AD-A197 648
 6.    Holford, R. Jr., Metal Finishing, Vol. 86, No. 7, July 1988, pp. 17-18.
 7.    Thrasher, H., Plating and Surface Finishing, February 1987, pp. 24-29
 8.    Personal Communication with Bill Sheets, Cadon Plating Company, Wyandotte
      Ml  48044.
 9.    Personaj Communication with Scott  Horoff, SPS Technologies, Jenkintown  PA
      and Plating and Surface Finishing, February 1986 on SPS Tech
 10.   Muraski, S.J., Machine Design, July  1989, pp. 95-98.
 11.   Donakowski, et al., Ford Motor Company, United States Patent #4,411  742
 12.   Ocean  City Research Corporation, Alexandria, VA, Results  presented at the
      quarterly MANTECH  meeting of MTL-AMC, Watertown, MA, November 7-8
      1990.
 13.   Brown, L,  Finishing,  Vol. 12, No. 11, December 1988, pp. 20.
 14.   Marce,  R.E., Plating and Surface Finishing, November 1982.
 15.   Horelick, P.O., Fourth EPA/AES Conference Advanced Pollution Control, Florida,
      1982.
 16.   Poll, G.H.,  Report on Reilly Plating, Brochure, Cadmium Council, New York,  NY

17.   Davies, Geoff, Finishing, July 1988, pp. 27-30.
                                     74

-------
          PARTICULATE AND HYDROCARBON EMISSIONS REDUCTION
                DURING WOOD VENEER DRYING OPERATIONS

                                     by

                                 Guy Lauziere
                           Production Machinery, Inc.
                               Bend, OR  97709

                                     and

                                  Jim Wilson
              Oregon State University, Forestry Products Department
                                 Corvallis, OR
ABSTRACT

      Production Machinery, Inc.'s project assessed if particulate emissions are
reduced or eliminated when wood veneer is dried using Radio Frequency (RF) energy
as compared to conventional energy sources. Current dryers use natural gas, wood
waste, or electric resistance heating in the direct or indirect heating of veneers. The
premise is that RF energy, attracted directly to the water in the wood, is able to heat
the water and drive it from the wood cells at lower temperatures than are required by
conventional dryers. Due to lower temperatures, there is little or no release of
hydrocarbons (smoke) from the wood compared to emissions generated from other
types of dryers.


INTRODUCTION

PROJECT DESCRIPTION

Outline of Process

      Production Machinery, Inc. investigated the particulate and hydrocarbon
emission rates from a wood veneer dryer that used Radio Frequency (RF) energy to
dry veneer.  The emissions, particulates, and hydrocarbons in the form of condensible
organics are a result of the veneer being subjected to process handling and elevated
temperatures.  Sliced wood veneer must be dried prior to its use in products such as
overlays, laminates, plywood, and laminated veneer lumber.  This drying is required
due to the moisture that is  introduced during the pre-conditioning process.  This
conditioning entails exposing the original flitches to either steam or hot water to
elevate the temperature and moisture content of the wood prior to slicing. Industry
experience has determined that a core temperature of approximately 140°F and a
moisture content of at least 40 percent will yield the highest quality slice.

      Conventional veneer dryers use hot air blown over individual sheets of veneer.
Heating occurs from the outside to the inside to dry the wood.  The  heat energy  is
commonly provided by steam heat exchangers (generated with  wood hogged fuel) or
are direct fired with wood hogged fuel or natural gas. The RF dryer provides internal


                                      75

-------
heating of the veneer combined with hot air to evaporate vacating moisture from the
veneer. This combination of heating methods should promote more efficient drying
practices. To establish the benefits of RF drying of veneer, an RF-type dryer (covered
by U.S. patent 5,162,629,  November 10, 1992) was compared to a natural gas-fired
conventional drying system located on the same plant site.

Unique Process Advantages

       Although  RF heating systems are commonly used in many industrial
applications, its use for drying wood products has been very limited. RF drying of
wood products has several advantages over other heating processes used in the
forest products industry.  Conventional veneer dryers use hot air blown over or at the
surface of the veneer, heating from the  outside to the inside to provide the necessary
heat to all of the veneer for drying.  When the frequency of the RF system is properly
selected, the majority of the RF radiated energy  is absorbed by the water molecules
retained in the wood. This energy vaporizes the water, which escapes through the
surface of the veneer. As the water concentration decreases, proportionately less RF
energy is absorbed in the  dryer areas.  This selective energy deposition  retards the
excessive drying of areas in the veneer  that are  reasonably dry yet allowing continues
heating and drying of localized wet spots.  In conventional drying systems (see Figure
1), where heat can be provided by either hot air  blown across the sheet  -- a
longitudinal dryer - or hot air blown perpendicular to the sheet - a j'et dryer -- the heat
is conducted through the surface of the veneer for drying. As some areas dry, they
still continue to be heated.  Conventional veneer dryers operate in temperature ranges
from 340° to over 400°F, with recent practices at the lower temperatures  to reduce
emissions.  (This new practice results in lower production rates.) Present methods of
drying, because of high temperatures, long drying times, and heating method lead to
over-drying or scorching of some areas of the veneer.  To reduce scorching, some of
the veneer is typically not fully dried. This requires an effort to sort out wet material
and return it to the dryer.   This wet veneer  is referred to as "redry."


Figure  1.     Diagram of Conventional Veneer Dryers
                                     longitudinal Dry«r
                         v«n»«r
                         to
                                     JrtDry*
                                      76

-------
      In an RF dryer, the drying methods are combined: the RF heats the internal
water in the veneer and the blown (forced) hot air reduces the surface layer insulating
the wood, thus providing heat to aid evaporation of water from the surface.  With the
RF dryer, the particulate and hydrocarbon emissions should be reduced due to lower
operating temperatures and reduced exposure times to elevated temperatures.  In
addition, a much higher percentage of veneer should be properly dried in a  single RF
dryer cycle, that the variance in moisture content within veneer slices should be
smaller, and damage to veneers should be reduced due to minimized handling.

Process Schematic

      The RF system combines a conventional jet dryer -- where hot air is directed at
the surface of the veneer through air caps - with an RF generator that transmits its
energy through a series of electrodes through which single sheets of veneer are
conveyed.  (See Figures 2 and 3.)  This RF configuration is referred to as a  stray field
system. The remainder of the RF dryer system consists of the support, drive, and
idler rolls for transporting the sheets of veneer through the dryer.
 Figure? 2.     RF Generator
 Figure 3.    RF Dryer
                                       77

-------
 APPLICATION

 Process Replaced

       During the drying of wood, water vapor, wood particulates, and volatile
 compounds are released into the atmosphere. In typical commercial wood veneer
 drying operations, 20 percent to more than 70 percent of the final dry wood weight will
 be evaporated as water.  The paniculate material is wood dust that is eroded from the
 surface of the veneer during handling.  A variety of low molecular weight organic
 compounds are also released along with the water vapor. The composition and
 amount of this vapor depends on the species of wood being dried and on the drying
 schedule or time-temperature history used in the drying process.  Low-temperature
 drying (air drying)  of many species is possible, which results in low release rates for
 these organic compounds. This drying method is  of little use in commercial
 operations because of the long drying times and poor process control. Elevated
 temperatures are used in forced hot air drying methods to speed water removal rates-
 however, this results in disproportionate release rates of volatile organic compounds. '
 In most veneer drying systems, heat is passed directly across or at the veneer
 causing high surface temperatures ranging from 340° to over 400°F.   These
 temperatures are high enough to pyrolize the  organic vapors and the entrained wood
 particulates that are transported out of the dryer's exhaust system. This partially
 combusted material not only results in the characteristic "blue haze" seen around
 veneer dryers but also generated highly combustable condensates that coat the inside
 of the dryer and the exhaust ductwork.

 Emissions Reduced

      The use of RF heating to dry veneer is expected to improve several  process
 considerations.  First,  particulate emissions should  be reduced because of shorter
 drying times that reduce handling.  Second, the production of partially combusted
 organic compounds should be lower because of reduced veneer temperature and
 drying time and the more efficient method of heating with combined RF and
 conventional forced hot air. The RF system should have significantly higher water
 extraction rates prior to causing degradation of the veneer.  Because  the RF is able to
 direct energy to the location of the moisture, overheating  of dry areas will be
 minimized and the  quantities of volatile organics, which are measured as condensible
 organics, will be significantly lowered.

      Another processing advantage of the RF system versus conventional dryers is
 that the final moisture distribution within a sheet of veneer and from sheet to sheet will
 have lower variance.  This results in a lower dryer reject rate  (redry rate) for the RF
 system, and as a result of reduced handling, particulate emissions are reduced.  This
 lowers operating costs, increases productivity, and  improves  product quality.

 Cross Segment Uses

      Although this study shows the potential  of the RF drying of sliced veneer for
 reducing emissions, there are other potential uses of this technology in the  forest
 products industry.  The technology could be adapted for drying rotary peeled veneer
that is used for making plywood, laminated veneer lumber, and overlays.  The RF
 portion of the equipment also could be used to preheat wood particles and wood


                                      78

-------
strands -- used in making composite panel products -- prior to drying in rotary
dryers. The RF used in this manner would reduce both particulate and hydrocarbon
emissions.  Other applications include preheating composite panel mats prior to hot
processing to improve productivity.


PROCEDURE

DEMONSTRATION PROCEDURE

      Four test runs were conducted, two with the RF dryer and two with the
conventional gas fired dryer, of one to two hours per run. Each system, the
conventional and the RF, used essentially the same data gathering methods and
materials.  Two wood veneer feedstocks were used in this study: one stock was
sliced, clear, vertical-grain Ponderosa Pine in strips with dimensions of 0.0833 x 5.00 x
84.25 inches; the second  stock was rotary peeled Douglas-fir veneer with dimensions
of 0.10 x 51.0 x 112.5 inches for the RF dryer and 0.10 x 27.0 x 112.5 inches for the
conventional gas dryer. In addition to particulates and hydrocarbon emissions
measurements, other process parameters included exhaust conditions, dryer
conditions, veneer production, veneer characteristics, and amount of moisture
removed from veneer.

      The drying equipment used in this study is located at Western Veneer and
Slicing, a forest products  production  facility White City, Oregon. This city is in the
Rogue River Valley, which is currently classified by the U.S. Environmental Protection
Agency (EPA) as a non-attainment air shed. Two separate production lines were used
to reduce  the moisture content of high quality sliced veneer from a wet moisture
content of approximately 40 percent  (oven-dry wood  basis) to a dry moisture content
of approximately 10 to 15 percent. Both production lines typically use similar
feedstocks and are physically  located inside a common building.

      The RF dryer consists of a single-pass conveyer with  an active drying length of
approximately 24 feet (see Figure 4). The usable conveyer width is 60 inches.  This
unit used a single RF generator with  a maximum rated output of 150 kilowatts
operating  at a nominal frequency of 3.8 megahertz. This heating  source is
supplemented by three natural gas forced draft  burners each supplying a maximum of
300,000 BTUs per hour.  This  unit used a single zone thermostat controller that was
set at a nominal temperature of 340°F for the study.  The conveyer can be set so that
the minimum residence time for veneer in the dryer is 42 seconds.  In production,
residence times are operator adjusted so that the targeted exit moisture content can
be achieved. Residence time  ranges from 50 to 90 seconds for the 0.08-inch thick
Ponderosa Pine veneer.  The system has a single exhaust fan that pulls air through
the conveyer's entrance and exit ports and vents via  sheet metal ductwork through the
roof of the main building.  This fan circulates approximately 1,500 ambient cubic feet
of air per minute to the environment.
                                      79

-------
Figure 4.     Diagram of RF Dryer System
                                             hotalrj**
      The conventionally heated veneer dryer is a single pass unit with an active
drying length of 58 feet with a usable width of 40 inches (see Figure 5)  This is
followed by a 13 foot forced air cool down zone. Heat is supplied by 4 forced  air
natural gas burners each having a maximum output of 300,000 BTUs per hour  The
system has a single exhaust fan and vent system that exhausts through the ceiling of
the mam building.  This system circulates approximately 1,150 ambient cubic feet of air
per minute.  The residence time to dry typical sliced Ponderosa Pine veneer (0 080-
inch thickness) is from 80 to 150 seconds.  This unit uses a three zone heat control
l^l?!7] thau ls^et by plant Per80™®1 for this study at an average dryer temperature of
340 F for the three  zones.
Figure 5.    Diagram of Conventional Gas Dryer System
veneer
In  —
                  vent
           111111
           t    t    T    t    T    t    t    t
                                                        vent
                                                                      veneer
                                                                      out
                      Drying
                      Section
                                                    Cooling
                                                    Section
                                     80

-------
EVALUATION PARAMETERS

      The measured parameters for the study are categorized and defined in the
following sections.  The experimental methods apply to both the RF dryer and to the
conventional natural gas fired dryer. Any differing methods are explained in the
appropriate section. To ensure accuracy, a majority of the data were recorded on a
Campbell CR21X data  logger and transferred to a personal computer for analysis.
Vent Stack Emissions
      The test method used for collecting the particulate and hydrocarbon emissions,
as well as the flow, temperature, and humidity of the exhaust, is defined in the Oregon
Department of Environmental Quality's (ODEQ) book of Industrial Regulations Method
7. An independent testing laboratory, BWR Associates of Medford, Oregon, a
company experienced with emissions studies and ODEQ Method 7, collected all vent
stack emissions in cooperation with Oregon State University personnel.

Natural Gas Use

      Natural gas flow rate was measured using an American Meter Company Model
A1-1000 pulse natural gas meter that was installed at each dryer to record total gas
use. These units contain  a totalizer and a pulsed switch closure that is counted by the
CR21X data logger. In addition, a switch closure device was placed on the burner
gas solenoids that recorded when the burners were on.  One burner was monitored
on the RF dryer since all burners were controlled by a single thermostat. On the gas
dryer, two burners were monitored since they were independently controlled.

      The composition of the natural gas determines the energy given off upon
combustion per a specified volume.  This value is critical in the determination of
energy use.  A value of 1,027 BTUs per cubic foot was obtained from the local gas
company in reference to Western Veneer and Slicing's natural gas. The gas
composition was estimated to be 75 percent methane and 25 percent butane.

      The composites of combusted gas, which include water, are critical in that they
are part of the vent stack emissions. The amount  of water contributed to the system
by  natural gas combustion was accounted for so that it would not be included as
water extracted from the wood.  No compensation was made for carbon dioxide or
carbon monoxide during combustion.

Exterior/Interior Dryer Ambient Conditions

      Wet bulb and dry  bulb ambient temperatures were determined for the exterior
ambient conditions with an Environmental Tectonics Corporation  Psychro-Dial model
CP-147 psychrometer at least once during each run.  The interior temperatures were
taken from thermocouples that were permanently mounted in the dryers. The
temperature and specific  moisture content values of the  air exhausting from the dryer
were obtained from BWR Associates' test results.
                                      81

-------
 Material Physical Parameters

       The main veneer parameters were the temperatures before and after the dryer,
 the weights before and after the dryer, and the moisture meter reading after the dryer.
 The targeted moisture content for material  exiting the dryer was 10 percent (oven-dry
 wood basis) for the sliced veneer and 5 percent for the rotary peeled veneer.  Of the
 total  material going through the dryer, 5 to 8 percent of the total number of veneers
 dried was sampled.  From this fraction, the moisture content and weights in and out of
 the dryer were recorded. Sampling 5 to 8 percent of the material yields a
 representative sample size.  Individual samples within the fractional sample consisted
 of a group of 4 veneer slices that were end-marked prior to the dryer to ensure that
 they  could be identified when exiting the dryer.

       Veneer moisture content measurements were taken after drying. Green end
 measurements were not taken because the veneer was above fiber saturation point
 and were off-scale on the moisture meters.  On-line moisture contents were measured
 with a hand-held, dielectric  moisture meter (Ward Systems, Inc.  Model 510) and
 recorded.  A 1 -inch thick piece of styrofoam was placed under each sliced veneer pair
 to isolate the reading.  Since this type of meter is dependent on wood thickness, two
 combined slices of the sliced veneer (a double thickness) were needed to obtain a
 correct reading. The recorded moisture content for the sliced veneer was an average
 of 6 readings within each sample group of 4 slices (3 readings from 2 veneers and 3
 from  the other 2 combined  veneers).  To ensure that the moisture meter was reading
 correctly, 7 to 8 samples (groups of 4) evenly spaced throughout the run  were
 measured with the moisture meter and weighed. These were taken to the laboratory
 to be oven dried to determine their actual moisture content.  A regression equation
 was developed that gave the true moisture content as a function of the moisture meter
 reading. Moisture content out of the dryer for the rotary peeled veneer was monitored
 by the same hand-held meter as the sliced veneer, but the values were not recorded.

       The sliced veneer's ambient temperature before the dryer was defined to be the
 same as the conditioning hot water bath temperature of 150°F. The temperature of
 the rotary veneer before the dryer was at ambient temperature since it had been
 stored in the plant for 1 or 2 days in a bundled unit  at room conditions. Since the
 dryer conditions remained fairly constant, only a small number of temperature
 measurements were made on individual pieces after each dryer run using a Scotch
 infrared temperature sensor.

       The samples of sliced veneer, from which moisture meter readings were taken
 (the 5 to 8 percent of total throughput), were also weighed before being run through
 the dryer using an OHAUS GT4100 scale and  immediately after the dryer with an
 OHAUS 1-10 digital scale. Due to the length of the veneer, a support (1.5 x 3.5 x 4 0
 feet of pine) was placed on  the scale and its weight  tared out before the veneer was
 weighed.  The exit moisture content, weight loss, and number of sheets dried were
 used  to determine the moisture loss for the sliced veneer.  Moisture loss for the rotary
 peeled veneer was determined by weighing the total amount of veneer before and
 after drying using a truck scale.  The difference was  the total water loss from drying.

      The total number of slices passing through the dryer during the run was
 recorded. Redry and fully dried material were tallied. Figure  6 illustrates how veneer
tallies were made in terms the types of veneer production  through the dryer as total


                                      82

-------
dryer throughput and total dry production.  This tally was done at the output end of
the dryer when the run was over.  The EPA uses the total throughput basis for its
calculations and considers the redry rate. However, PMI felt that a truer measure of a
plant's production would be the total dry basis, so both basis were used. On  the RF
dryer, the redry material was sent back through the system, so a running tally was
kept. On the conventional dryer, the redry material was kept separate and not run
back through the system  during the run.


Figure 6.    Flow Diagram of How Veneer Tallies Were Made
                                      Redry
                        Qreen .
                        Veneer
RF
Dryer
Dried
Production
(dry only)
                                         Total
                                         Throughput
                                         (dry+redry)
                                                Redry
                        Qreen _
                        Veneer
Gas
Dryer
 Dried
 Production
 (dry only)
                                          Total
                                          Throughput
                                          (dry 4-redry)
       The length, width, and thickness were recorded from the veneer samples
 brought back to the laboratory to be oven dried (48 to 52 pieces from each run).
 Their length and width were measured with a Stanley 25-foot tape measure.  The
 thickness was determined with a Starrett hand-held micrometer, accuracy to 0.01
 millimeter. The number of sliced and rotary peeled veneer sheets dried were
 conducted at the end of each run. No distinction was made between dry and redry
 for rotary peeled veneer. All material passing through the dryer was considered to be
 fully dried material.

       The criteria for wood  quality included a visual  inspection of the same 5 to 8
 percent that was sampled for weight and  moisture content. The visual inspection
 involved discoloration and appearance  of cracks or other drying related defects
 apparent in the wood.
                                       83

-------
 ASSESSMENT SUMMARY

       While the project showed that the RF dryer could reduce emissions for sliced
 veneer -- the type of veneer the equipment was designed to dry -- care must be taken
 in using the results of this study beyond its scope since only one trial run was made.
 Although the RF dryer was able to dry the rotary veneer, it could not do so in an
 efficient manner during the test. The RF dryer was calibrated specifically for sliced
 veneer, and it is expected that a recalibration of the dryer for the rotary peeled veneer
 would improve the overall performance.

 COST OF DEMONSTRATION

       The total cost of the project was $27,727.87.  EPA contributed $21,652.87; PMI
 contributed $1,075; and Pacific Power and Light donated $5,000 towards the study.


 RESULTS AND DISCUSSION

 PERFORMANCE RESULTS

       The RF dryer reduced particulate emissions by 40 percent and condensible
 organics by 23 percent.  Condensible inorganics (that occurred at near background
 levels) were not  reduced for the sliced veneer drying when considering total dried
 production (see Table 1).  When the veneer tally is based on total veneer throughput
 the emissions reduction is less, 18 percent for particulate and 0 percent for both the
 organic and inorganic. The particulate component on a weight basis represents
 approximately 84 percent of the total emissions measured.
TABLE 1.    Summary of Emissions Reduction for RF Drying of Sliced Veneer
Emission Component
Particulate
Organic
Inorganic
Total Emissions
Reduction in Emissions
(%) Based on Total
Dryer Throughput
18
0
0
16
Reduction in Emissions
(%) Based on Totai
Dried Veneer
40
23
0
38
      The run of rotary veneer through the RF dryer was unsuccessful in reducing
emissions.  The equipment was able to dry the veneer, but because the "tuning" of the
RF generator was specifically tailored to the sliced veneer, the RF generator would not
load properly, providing only about 25 percent of its expected output power  If the
generator were tuned to full size sheets of veneer, it is expected that the RF dryer
would also reduce emissions for the rotary veneer as it did for the sliced veneer
                                     84

-------
PRODUCT QUALITY VARIANCE

      The RF dried sliced veneer was "brighter and less brittle" than the veneer dried
in the conventional dryer, thus better maintaining the quality of the veneer. This
observation could not be quantified, but was made by the experienced veneer graders
who were working on the production line at Western Veneer and Slicing.

CONDITIONS THAT IMPACT PERFORMANCE

      A number of production parameters affect the operating efficiency of the RF
dryer, and only those conditions reported in this study were evaluated. Such
conditions that impact dryer performance are: material, environmental, and equipment
parameters.  Material  parameters are percentage of energy contribution by RF and
conventional heating,  tuning of the RF, exhaust flow, temperature setting, and
residence time.  Environmental parameters include temperature and humidity.
Although not studied, equipment could be redesigned once the specific operating
conditions are defined to provide similar, if not greater reductions in emissions when
drying veneer.  For example, the performance of the RF dryer could be enhanced
further by reducing the amount of exhaust flow. This would improve energy  efficiency
and reduce natural gas combustion, which in turn would reduce hydrocarbon and
water vapor emissions.

TABULATION OF DATA

      Comparison of emissions  between the RF and conventional dryers was
performed under the following parameters: particulate  emission; condensible organic
and Inorganic emission; and water vapor emission.

      In  order to place both drying systems on an equal material basis, two sets of
computations were performed. These results present  the emission rates on  two
different basis: (1) emission rates relative to the total volume of material exiting from
the dryer ~ total throughput of veneer; and (2) emission rates relative to the  volume of
finished dry veneer product -- total dry production.  The EPA uses the total throughput
basjs  at a given redry rate; if the rate changes, the dryer needs to be recertified.  Both
basis  provide useful information about dryer efficiency, and  PMI has provided data in
the two common forms seen in the literature.  The emission rates relative to the total
amount of dried production for the full production cycle appears to be the more
relevant.  The two sets of emission data are provided in Tables 2 and 3.
TABLE 2.    Summary of Particulate amd Hydrocarbon Emission Rates
            [(pounds/(hour * MSF 3/8 inch basis)]
Emission
Component

Particulate
Run 1
RF Dryer with
Sliced Veneer
(A) (B)
0.124 0.135
Run 2
RF Dryer with
Rotary Veneer
(A)
0.189
Run 3
Conventional
Dryer with Sliced
Veneer
(A) (B)
0.152 0.226
Run 4
Conventional
Dryer with
Rotary Veneer
(A)
0.079
                                      85

-------
Condensible
Organic
Condensible
Inorganic
Total Emissions
(1)
Total Water
Vapor (2)
0.021 0.023
0.002 0.002
0.147 0.160
179 196
0.050
0.006
0.245
408
0.020 0.030
0.002 0.003
0.174 0.259
195 290
0.022
0.003
0.104
214
      .uuu square reel (surrace measure) ot veneer at a nominal thickness of 3/8 inch; 1 MSF of veneer
 equals 31.25 cubic feet of wood
 A) Based on total veneer through dryer
 B) Based on total dry veneer through dryer (total-redry)
 1) Paniculate + Condensible organic + Condensible inorganic
 2) Wood extract + Combustion products + Make up air
TABLE 3.    Specific Release Rates (pounds/MSF)
Release Rate

Paniculate
Condensible
Organic
Condensible
Inorganic
Total
Water Vapor
Run 1
RF Dryer with
Sliced Veneer
(A) (B)
0.178 0.195
0.031 0.033
0.003 0.003
0.209 0.241
258 282
Run 2
RF Dryer with
Rotary Veneer

0.299
0.079
0.010
0.388
408
Run 3
Conventional
Dryer with Sliced
Veneer
(A) (B)
0.212 0.316
0.028 0.042
0.003 0.005
0.243 0.363
273 406
Run 4
Conventional
Dryer with
Rotary Veneer

0.111
0.030
0.003
0.144

(A) Based on total veneer through dryer
(B) Based on total dry veneer through dryer (total-redry)
• Emission rates are pounds per hour
• All production runs are assumed to be done at typical production throughput rates as given in Table 1
       As can be seen in Table 2, the participates release rates for sliced veneers have
a higher level for the conventional dryer than for the RF dryer (run 1 versus run 3) for
both tally basis computations.  As the feedstocks for both runs were identical, no
material differences should have caused this difference. The RF particulate emission
rates are  18 and 40 percent lower, based on total throughput and dried production
respectively, than the conventional dryer. These values are well above the
experimental uncertainty. A  significant reduction of 23 percent exists for Condensible
organics (hydrocarbons) on  the total dried production basis, no difference  occurred on
a total throughput basis.  Release rates  for inorganic material trapped in the
Condensible organic fraction is a very small part of the total release for all runs.  These
values are near the minimum measurable levels for this test.
                                        86

-------
      For the rotary peeled veneer (run 2) dried with the RF drier, the RF system
failed to load properly and is therefore not representative of what this technology can
provide.  This run is an example of a poorly implemented conventional veneer dryer,
since the output of the RF was much lower than its rated capacity, resulting in a larger
contribution  of natural gas heating. Excessive local heating was observed near knots.
Some knots were severely discolored or dislodged during the drying process. A
change in dwell from 6 to 4  minutes during the run eliminated much of this problem.
As a result, significant amounts of wood debris were generated and entrained in the
circulation air and ejected from the system. Run 2 also produced approximately a
factor of 2 higher levels of condensible organic release than the other 3 runs. This is
consistent with the harsh heating observed near knots.

      The release of water vapor from the dryer was also assessed. Although water
vapor is not considered an environmentally sensitive material, it is what most casual
observers see  emanating from veneer dryers. These data show a significantly
elevated release from Run 2 as compared to the other 3 runs.  This is due largely to
the very low production rate for this trial.  The longer the veneer residence time in the
system, the  larger the water vapor contribution made by the natural gas burners and
the makeup air stream relative to the amount of water contributed by the wood.  Water
vapor generated by the burners plus the water vapor present in the makeup circulation
air stream contribute 25 to 40 percent of the total emitted water vapor in these trials.  If
one computes water emission rates assuming that all 4 runs were done at the
maximum throughput rates for the material being dried and then compensate for the
water vapor added by natural gas combustion and external  makeup air for Runs 1
through 4 respectively is 7.08, 7.15, 7.63, and 7.33 pounds per square foot.

      The particulate emissions rate for rotary peeled veneer should be measurably
higher than the sliced veneer because of the dramatic differences in surface texture
between the two veneering operations. Data presented in Tables 2 and 3 for Runs 3
and 4 indicate  that the particulate emissions for the sliced veneer were between a
factor of two to three times higher than the rotary  peeled veneer.  Possibly, the large
fragments seen with the rotary veneer do not  get entrained in the exhaust air of the
conventional gas fired dryer and thus do not appear in the test samples.  This may
also be influenced by the width of the veneer relative to the  design width of the dryer's
conveyer system that would hamper vertical air  flow.  Although the exhaust rates for
both dryers  are similar, the internal volume of the conventional dryer is much larger
than the RF system, which may play a role in  the entrainment of particulates in the
exhaust air stream.

COMPARISON OF STUDY DRYERS TO OTHER COMMERCIAL DRYERS

      The emissions from these two veneer dryers can be compared with those from
other installations. Information provided by the  ODEQ on veneer dryers for two
softwood plywood plants appear in Table 4.  These data are for installations that use
rotary peeled Douglas Fir sheets similar to those dried in this study.  No particulate
emission data have been found for sliced veneer.  Volatile releases from sliced and
rotary peeled veneer are assumed to be the same and depend on the wood species,
temperature, and dwell.  Ponderosa pine, because of its chemistry, would be expected
to have higher values of organic emissions. For the Site 1 a data, the dryer was out of
compliance for opacity  (smoke) and required production adjustments. Site 1 b data
are more representative of a compliant system.  For the compliant tests the particulate


                                      87

-------
emission levels ranged from 0.09 to 0.18 pounds/(hour*MSF 3/8 inch) basis.
Condensible organic rates for PMI's tests fall in the same range as the results of the
ODEQ test data.
TABLE 4.    Summary of Paniculate and Hydrocarbon Emission from Other Veneer
            Drying Operataion Compared to thi Study on a Total Throughput Basis
Emission
Component

Paniculate
Condensible
organic
Condensible
inorqanic
Total (2)
Water vapor
(3)
Wood
species
Site la
non
compliant
0.827
0.168
0.0196
1.01
790
Douglas
fir
sapwood
Site ib
compliant
0.0915
0.0249
0.0018
0.118
449
Douglas
fir
sapwood
Site 2
compliant
0.177
0.0128
0.0026
0.192
329
Douglas
fir
sapwood
Run 1
compliant
0.124
0.021
0.002
0.147
179
Ponderos
a pine
Run 3

0.152
0.020
0.002
0.174
195
Ponderos
a pine
Run 4

0.079
0.022
0.003
0.104
214
Douglas
fir

1 1) Emission rates are: lbs/(hr* MSF 3/8 inch) basis
2j All production runs are assumed to be done at typical production throughput rates
3) Paniculate + Condensible organics + Condensible inorganics
4) Wood extract + Combustion products + Makeup air
5) Site 1a: 8.81 MSF/hour, test run #1, 0.10 inch thick sap; this run was smoking heavily and was out
of ODEQ levels
(6) Site 1b: 5.362 MSF/hour, test runs #2 and #3, 0.228 inch thick; these runs were within permitted
levels after adjustment from Run #1
(7) Site #2: 5.689 MSF/hour, 0.10 inch thick
      The ODEQ standard for opacity was not done during the project.  Although not
documented in the literature whether there is a relationship between the emission
concentrations measured as particulate and condensibles and the opacity
measurements, PMI suspects there is a positive correlation between the product of
concentration, the highest dryer internal temperature, and the amount of time the
particulates and organics remain at high temperature.  The lower residence time,
coupled with lower peak temperatures of the veneer in the RF dryer, may lead to
significantly reduced smoke generation. Secondary evidence to support this idea can
been seen in the significantly reduced internal  coatings of extractives in the RF dryer
compared to the conventional dryer.

COST/BENEFIT ANALYSIS

      Higher production efficiency can be realized when using RF technology in
veneer drying.  For example, the RF dryer processed 636 square feet of veneer per
                                     88

-------
hour more than the conventional dryer during this test (based on total dry production).
Assuming approximately $0.50 per square foot sale value, the RF dryer was producing
$318.00 of veneer per hour in excess of the conventional dryer. This amounts to over
$600,000 in additional sales annually.

      The RF project illustrated a number of benefits to industry.  Some are difficult to
quantify, as they pertain to variables such as the environment, or they are based on
matters of opinion, such as aesthetic veneer quality.  RF drying seems to be more
efficient compared to conventional drying. The natural gas savings for the RF dryer
obviously is offset by electrical consumption, but considering actual dry veneer
production, the  RF dryer consumed approximately 38 BTUs per hour per square foot
of veneer less than the conventional dryer in this test.  The conventional dryer also
generated approximately 23 percent more redry material than the RF dryer, which
equates to more run time and energy usage.

      The RF dryer requires only about half the space of an equivalent conventional
machine. This allows the user to further optimize utilization of the floor space required
for plant operation.  Due to cleaner emissions, the maintenance requirements for the
RF dryer are far less  demanding than its conventional counterpart primarily because of
the reduced levels of pitch and resin buildup within the dryer.

      RF drying also proves to be advantageous when evaluating the residual
moisture distribution after the drying process is complete.  Wet spots can be a
problem for a conventional dryer, whereas in RF drying, a very  uniform, normalized
product is produced.  RF energy is attracted to any contained moisture within the
veneer and works only on those areas, without affecting zones  that are already dry.


CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      RF drying technology  has a great deal to offer in today's wood products
industry marketplace.  Considering natural resource shortages and environmental
issues affecting it's producers, the benefits associated with dielectric heating will have
an impact.  In an attempt to do more with less, there has been a major trend towards
the production of thinner and thinner veneer products.  As mentioned previously, all of
this veneer will have to be dried as part of the process. RF drying will provide one of
the safest and efficient methods of handling this valuable resource with the least
amount of impact on the environment. RF drying methods can aid manufacturers in
keeping up with the demand for veneer while remaining competitive and can eliminate
cost overruns created by excessive redry runtime and compromised veneer quality.

Phase of Development

      The Dry-Tech Radio Frequency Veneer Dryer,  U.S.  Patent #5,162,629, is
currently in production status. To  date, two machines have been built, shipped, and
installed for the same customer in  White City, Oregon.  PMI is responding to
considerable  interest in the RF dryer, including a number of international inquiries.


                                      89

-------
Further development is being considered to modify this technique for use in the
plywood industry.  Redry and wet spots are ongoing problems for plywood
applications that could be rectified through the use of RF drying.

Target Industry Potential

      This technology could be easily adapted for drying of rotary peeled veneer,
which is used for making plywood, laminated veneer lumber, and overlays. The RF
portion of the equipment could also be used to preheat wood particles and wood
strands - used in making composite panel products -- prior to their being dried in
rotary dryers.  Other applications include preheating composite panel mats for
hardboard, particleboard, medium density fiberboard, and oriented strand board prior
to hot pressing to improve productivity.
                                     90

-------
                    CONDUCTIVE POLYMER COMPOSITES
          TO REPLACE HEAVY METALS IN COATINGS AND ADHESIVES

                                     by

                        Harry S. Katz and Radha Agarwal
                         Utility Development Corporation
                             Livingston, NJ  07039
ABSTRACT

      Improved polymer matrix conductive coatings and adhesives are needed to
replace current products that use heavy metals.  Utility Development Corporation
developed a low-cost polymer adhesive that has equivalent conductivity to those
currently available and assessed its physical and electrical properties.  The project
focused on the inclusion of short graphite fibers, carbon/graphite microspheres, and
conductive carbon powder fillers as the conductive elements.


INTRODUCTION

PROJECT DESCRIPTION

      Utility Development Corporation developed and produced small quantities of
conductive polymer composites as substitutes for heavy metal solders and other
metallic  components, such as lead, tin/lead, and silver, in electronic applications. The
main formulation contains carbon/graphite fibers and carbon/graphite fillers in a
thermoplastic matrix.  Most formulations were extruded  in a one-eighth inch diameter
rod that could be used in the same manner as conventional solder.

      In addition, Utility Development formulated and tested a number of conductive
polymer adhesives and coatings.  In these studies, the carbon/graphite fibers and
fillers were dispersed in water-based urethane and in epoxy resins.

      The project focused on the formulation and experimental testing of the
conductive adhesives. As Utility Development has limited experience in the electronics
field, this program did not accent specific end uses.

Unique Product Features/Advantages

      The low-cost and environmentally safe conductive fillers should provide high
performance for the many applications usually requiring the use of a lead-based
solder.  In conventional soldering, flux and solvent are required to obtain a clean
surface  and good bond strength. The use of polymer matrix conductive composites
do not require fluxes or solvents.
                                      91

-------
Process Schematic

    The process schematic for polymer matrix conductive products is shown in
Figure 1.

Figure 1.   Polymer Matrix Conductive Products
   CONDUCTIVE
     FILLERS
    (graphite fiber
     and filters)
    COATINGS
                          MIXING
EXTRUDE
  RODS
                   POLYMERIC
                   MATERIALS
                  (polyamide/epoxy/
                 water based urethane)
ADHESIVES
                        TESTING
                           92

-------
APPLICATION

Process/Products Replaced

      Conductive polymer matrix solders, coatings, and adhesives were formulated to
replace current products that use heavy metals such as lead, tin/lead, and silver.
These metals, which are used extensively in electronic devices and automotive control
systems, are  potential toxic contaminants to the environment.

Wastes Prevented

      The conductive materials replace products containing heavy metals, such as
lead, tin/lead, and silver.  Fluxes and solvents (e.g., methyl ethyl ketone, toluene) used
in a conventional soldering process are also eliminated.

Cross-Segment Uses

      Conductive adhesives may be used where a convenient, low temperature-curing
conductive material is desired. These conductive composites may be used by the
electronics industry to replace lead and tin/lead solders.  They also may be used as
coatings, antenna, or circuitry on "smart" credit cards and electromagnetic shielding.


PROCEDURE

DEMONSTRATION

      Polymer matrix solder materials, conductive adhesives, and coatings were
produced by  using a high loading of conductive fillers in various polymer matrices.
Most of the work was conducted with different types of carbon/graphite fibers, flakes,
and powders at various loading levels in order to determine the proper  materials and
formulation for a highly conductive polymer composite.

      Utility Development assessed three grades of Union Camp's polyamide resin:
EPI REZ 2641-D, EPI REZ 2643-D, and EPI REZ 2645-D for use as the  matrix of the
conductive composites.  The melting point of these resins ranged from  120° to 140°
centigrade.

      A two-component thermoset eppxy resin system was also evaluated. This
system consisted of Shell Epon 826 with Henkel Versamid 140 as the curative.  Both
components were highly loaded  with conductive fillers before mixing and curing.

      Conductive fillers that were evaluated included ten different grades of graphite
powders and a special grade of  very high electrically conductive graphite filaments.
Representative products included P-120 Graphite Filaments (AMOCO Performance
Products), CABOT-XC72R  Graphite Powder, and M201S Milled Graphite Fiber (Kreha
Corporation). The filaments were chopped or broken during mixing procedures to
obtain a very low aspect ratio.
                                      93

-------
       A mini-mixer injection-molder quickly mixed and molded many different
 permutations of fillers.  The temperature and mixing cycle were varied for the different
 formulations. The resultant product was a continuous extruded rod, approximately
 one-eighth inch in diameter, and looked like standard commercial solder, except for
 the black color.

       In the beginning of the project, the conductivity levels of the formulations
 containing P-120  were not consistent. This graphite fiber is sold in continuous
 multifilament spools. In early experiments, filaments were broken into irregular lengths
 using the Mini-Max Molder to obtain a low aspect ratio for the flexible rod-type solder
 product.  Since one factor for erratic conductivity levels could be attributed to the
 uncontrolled aspect ratio pf the graphite fibers, Utility Development elected to ball-mill
 and screen-separate the filaments, selecting a narrow length  range.  After the P-120
 fiber was ball-milled, the broken fragments were screened through 50,  100, and 200
 mesh screens.  +50 and -200 mesh fibers were discarded, and -50 and +100 mesh
 fibers -were used.  More consistent conductivity measurements were obtained using
 the ball-milled fibers.

       To achieve the maximum conductivity, very high loadings of conductive fillers
 are required in  the polymer (one part resin to at least two parts filler by weight).  At
 these high levels  of fillers, the adhesion to the substrates  becomes very poor.  Utility
 Development evaluated different resins and coupling agents,  such as silanes and
 titanates, to improve adhesion,  but were not initially successful. After the preliminary
 fiber screening  tests, Utility Development investigated proper  filler packing concepts
 and selected the  best combinations and ratios of fillers and fibers.  This minimized the
 problems of poor adhesion to the substrates.  Various ratios  were tested to optimize
 the combinations in accordance with packing concepts to achieve good adhesion,
 maximum packing, and high electrical conductivity.

 Packing Concepts

       Packing  concepts  are vital to the success of the polymer composite. All
 conventional polymers are good electrical insulators.  In order to obtain a polymer
 composite/adhesive with extremely high electrical conductivity, filler loading in the
 polymer must be  maximized.  If packing concepts are  not used, a very high loading of
 filler results in a composite with very poor physical properties and a high void content
 due to processing problems with resin-starved polymer solder.

       Examples of the effectiveness of good packing combinations are illustrated in
 Figures 2 and 3, which are taken from the Handbook of Reinforcements for Plastics,
 published in 1987 by Van Nostrand Reinhold.  (Mr. Katz, the principal investigator of
this project was co-editor and contributing author of this book.)
The co-editor of the Handbook, Dr. John V. Milewski, was the main proponent of the
development of packing concepts.
                                       94

-------
Figure 2.
Addition of Small Spheres to Large Spheres
                    1.0 LARGE
                    0.0 SMALL
                   DENSITY 62.5%
               0.85 LARGE
               0.15 SMALL
              DENSITY 72.0%
   0.72 LARGE
   0.28 SMALL
DENSITY 85.0% MAX,
Figure 3.
Addition of Large Spheres to Small Spheres
                     0.0 LARGE
                     1.0 SMALL
                   DENSITY 62.5%
                0.33 LARGE
                0.67 SMALL
               DENSITY 72.0%
    0.72 LARGE
    0.28 SMALL
 DENSITY 85.0% MAX,
       Figures 2 and 3 illustrate the densification that occurs when small spheres are
added to large spheres.  Maximum density is obtained when the small spheres are
packed to their maximum density within the voids of the larger spheres.  Each
illustration represents the same volume of solid material; thus, the relative bulk volume
decreases as densification occurs. In Figure 2, in each step towards greater density,
a large sphere  is removed and the same amount of solid material is replaced by small
spheres within the voids of the remaining larger spheres. Figure 3  illustrates how
densification occurs by the opposite process, in which a number of small spheres and
their associated voids are removed, and the same amount of  material is replaced as
one large solid sphere.
                                       95

-------
       The theoretical packing curve for two different size spheres is shown in Figure
 4.  The solid line is a theoretical packing curve for an infinite size ratio R = «   For this
 example, the ratio  R is the diameter of the large sphere divided by the diameter of the
 small spheres.  The maximum density point represents the condition illustrated in
 hgures 2 and 3 at the extreme right.  In Figure 4, the composition of the mixture is
 shown by the horizontal scale X, the volume fraction of the small spheres  for a total
 volume of unity. The  left-hand ordinate is the relative bulk volume and is defined such
 that 1.0 is equal to 100 percent solid material (100 divided by the percent theoretical
 density).  A material with a relative bulk volume of 1.6 would be 62.5 percent
 theoretically dense. Thus C gives the experimentally determined packed volume of the
 large spheres, and M  gives that of the small spheres.  By using relative bulk volume
 rather than percent theoretical density, the packing curves for the infinite size ratio
 becomes two straight lines.


 Figure 4.     Theoretical Packing of Two Sphere Systems
          2
                     MALL SPHERES
                     FILLING
                     VOIDS
LARGE SPHERES REPLACING
MANY SMALL SPHERES AND
THEIR VOIDS
            1.6 C
                « = 1
                » = 0

            LARGE SPHERES
COMPOSITION
 (« * y = 1)
  « - 0
  y - I
SMALL SPHERES
       A proper ratio of the diameters of two or three sizes of fibers and microspheres
results in good interstitial packing and a minimum void content  Under these
conditions  there will be a lower resin demand to fill the interstitial voids and coat each
filler particle so that adequate flow is maintained during the molding procedure   Also
the molded part will have less air pockets or microvoids that reduce the physical and'
electrical properties. In contrast, if sphere ratios or particle size distribution is wronq
the result will be poor flow, high void content, high shrinkage, high electrical
resistance, and low physical properties.

       For conductive composites, coatings, and  adhesives, 1 part of resin to 1 5 parts
of a combination of fillers was found to be an optimum ratio to obtain good adhesion
and conductivity. Also, 3 parts of the ball-milled carbon fiber  to 1 part of carbon
powder yielded a good packing combination
                                       96

-------
EVALUATION PARAMETERS

      Dr. Haim Grebel, Professor at New Jersey Institute of Technology, supervised
conductivity tests at NJIT.  The procedures are described below.

Use of Conductive Polymer Composite as Solder for Metals

      Aluminum was selected as the metal to be bonded. It was machined into two
plates, 1 x 1  x 0.078 inches (25.4 x 25.4 x 2 millimeters).  The conductive polymer
composite samples containing resin, graphite fibers, and powder were cut into beads,
each of equal volume (10.5 cubic millimeters).  The samples were in the form of rods,
and the volume of the beads was estimated assuming they were perfect cylinders.

      The diameter of the beads was measured using a micrometer, and the length of
the bead was cut to  provide the same volume in each sample.  In order to obtain the
same thickness of the bonding material between the plates, cover slips, 0.0085 inches
thick, were used as shown in Figure 5.


Figure 5.    Experimental Configuration for Testing the Soldering of Polymeric
            Materials
                          ALUMINUM PLATE
COVER SLIP
                                                               COVER SLIP
                                                      IT" ALUMINUM PLATE
                                                      BONDING PLASTIC
      Etoth the plates and the plastic bead (positioned on one of the plates) were
heated on a hot plate to the point where the bead became tacky. The second plate
was then placed on the plate holding the bead.  This "sandwich" was removed from
the hot plate and pressed together for approximately 30 seconds and allowed to cool.
                                     97

-------
       The temperature of the hot plate was monitored using a thermocouple  The
 cover slips were removed and the resistance bond measured at direct current (dc)
 and at 1 kilohertz. A control measurement was made using tin/lead solder  The error
 in resistance values was estimated at 7 percent; this was based on the possible error
 in the volume of solder bead.
 Transmission Characteristics of a Slotline Made of Conductive Polymer Adhesive<
 Microwave Frequencies                           ~~                      *
at
       A film was cast on a glass slide of a formulation containing water-based
 polyurethane and graphite filler. This conductive polymer film was pressed between
 two heated microscope slides, 2 x 1 x 0.039 inches, to form a thick film.  Prior to
 deposition, the glass slide was cleaned sequentially with acetone, methanol  and
 distilled water to remove particles that would alter the conductivity of the film  To
 create a slotline, the film was ablated longitudinally at its center using a high power
 ultraviolet laser. A pair of coaxial cables were connected to the slotline bv pressina
 the former against the latter, as shown in Figure 6.


 Figure 6.     Experimental Configuration to Test High Frequency Response of the
             Filled Polymers
                             TO NETWORK ANALYSER

                               7"
                                  /
                                   COAXIAL CABLE


                           GRAPHITE LOADED PLASTIC
                   GRAPHITE LOADED PLASTIC
             GLASS SLIDE
      Pieces of low-resistance conductive polymer composite thick film (UDC 24R)
were placed between the coaxial line and the slotline for better contact  An HP
Lightwave Component Analyzer was used as a network analyzer to determine the S-
parameters of the line, and an HP 7475A plotter was used to plot the characteristic
L/ur v"s.
                                     98

-------
      The resistance between two close points on the film was found to be 30 ohm
(as measured by a multimeter).  The contact resistance between the film and the
coaxial line was also found to be 30 ohm. Thicker films had lower resistance, better
transmission characteristics,  and produced a lower mismatch at the transition points
compared to the thinner films (of the order of tens of m  that were developed by
spinning over glass slides).


COST OF DEMONSTRATION

      The cost of this demonstration project was $59,337. EPA provided $25,000
through the Pollution Prevention By and For Small Business Grant Program, and Utility
Development contributed $34,337.


RESULTS AND DISCUSSION

PERFORMANCE RESULTS

      Results indicate that the conductivity of the plastic materials filled with graphite
fibers and powder may be suitable for many soldering applications.  These materials
need to be tested in actual applications, and additional research may be necessary to
refine the product.

PRODUCT QUALITY VARIANCE

      Product quality variance depends on such factors as method of application,
thickness of the coating or solder joint, temperature, and mixing time and speed of the
formulation.  All of these factors can be varied within a wide range after first
establishing  product and process specifications.

CONDITIONS THAT IMPACT PERFORMANCE

      The uniformity of dispersion of the conductive fibers and fillers is one key
performance factor of the conductive polymer systems. Another condition that must
be considered is the presence of moisture in the polymer solders and coatings, either
before or after application. One of the primary matrix  materials tested was a series of
polyamide resins that have excellent adhesion,  physical properties, and application
characteristics. However, they tend to adsorb moisture. These solders and coatings
should be completely dry before conductivity measurements are taken.


TABULATION OF DATA

Conductive Polymer Composite as Solder for Metals

      Bulk resistance of the aluminum plates, measured with a multimeter, was 0.2
ohm. Contact resistance between the multimeter probes and the plate measured 0.1
ohm.
                                      99

-------
        The resistance observed at dc and at 1 kHz was practically identical  The
 actual electrical resistance measurements of various soldering materials are given in
 lable 1   As can be seen from the Table, some filled polymer solder materials possess
 relatively low resistance values, especially UDC 24, as compared to the control metal
 solder.

 TABLE 1.    Electrical Resistance of Filled Polymer Materials
Filled Polymer Sample
Tested
UDC 23R
UDC 24R
UDC 26R
UDC 27R
UDC 28R
UDC 29R
volume of sample
10.5 mm3
10.5 mm3
10.5 mm3
10.5 mm3
10.5 mm3
10.5 mm3
Electrical Resistance
(ohm} *
24, 56
8, 1
75, 70
50, 28, 37
70, 45, 63
47, 77
Control Metal Solder
Pb/Sn: 50/50 bv weiaht
Pb/Sn: 60/40 bv weiaht


.6

                                          P°°r'y 3dhereS l° many Other materials including many
 ™™r              conta'ns two parts:  the resistance of the contact (the interface between the filled
 polymer and the aluminum plate) and the bulk resistance of the filled polymer Itself  Most of the
 [SSSS?59   t6-8  °Wn In the .Tabl? are attributed to the interface resistance.  That was confirmed by
 using longer strings of polymeric solder material.  Utility Development concluded that contact resistance
 SrX^SL** ?ma" vol.ume °f Pdymerte beads.  This"explains the large variation in SSS5SS2?
 S nSnnmintaKhrSIOn ^/a*8"** ™y result in variations in resistance values.  Nevertheless,
 Utility Development believes that polymeric soldering materials are promising for the following reasons:

       They exhibit relatively low resistance values that may be tolerated by the electronics industry.

•      They are environmentally safe.

       The contact resistance problem can be solved by using other polymer materials or couplinq
       Kfh £ S?ld®nn9 materials are usually matched to the electronic circuit materials even though
       ^Hnet™ Clrcuit and t.1?8 soldering material are good conductors. For example, the electronics
       industry uses silver alloy as soldering material to match the silver-made  circuitry layout.


Transmission Characteristics of a Slotline Made of Conductive PnK/mpr Adhesives at
Microwave Frequencies                              "-*	s^-a-


       The S-parameter plots for the  slotline may be seen  in Figure 7  The
transmission  coefficient (S21)  is greater than -10 decibels for frequencies up to 1
gigahertz (the curves were plotted over the range of 0.3 megahertz to 3 gigahertz)
                                          100

-------
Figure 7.     Voltage Transmission as a Function of Frequency
      N
      O
                                                            <
                                                                      N
                                                                      I
                                                                      O
                                                                      O
                                                                      O


                                                                      O
                                                                      O
                                                                      O
                                                                        •

                                                                      O
                                                                      O
                                                                      O


                                                                      n


                                                                      n
                                                                      o
  QJ
  01
      o
      •e-
      tn
  m
  i

  IL
  UJ
  oc
  m
  u

  ai
  (D
  <



  O
  O
  OJ


  Tl
  I
  u
   UJ
(M \
U UJ
                                                                       N
                                                                       I
                                                                      O
                                                                      o
                                                                      o

                                                                      o
                                                                      o
                                                                      tn
                                                                       rx
                                                                       <
                                                                       H
                                                                       0)
                                         101

-------
       Suitable conductive plastics as a substitute for metal components have great
 potential for use in microwave integrated circuits.  The application of these trials on
 actual circuit boards was not performed.

       Utility Development was successful in obtaining relatively low electrical
 resistance,  ranging from 0.2 to 20 ohms, with conductive fibers and fillers in polymer
 matrix composites. Even though this resistance is not as low,  about 0 5 to 1 0 ohm
 as the heavy metal solder materials, there should  be many applications for this
 performance level, and future work will provide improvements in further resistance level
 reductions.
   ^ u-     maJ°r advanta9e of composite polymers is that they are conductive at low
 and high (up to 1 gigahertz) frequencies. Several devices may now be realized in
 plastic materials: short connections between wires, plastic antennas and high
 frequency circuitry  The dispersion of the carbon/graphite fillers inside the polymeric
 matrix dictates the low and high conductive properties of the overall material  This can
 be achieved by using the same polymeric matrix by varying the amount of conductive
 filler in that matrix.  Dr. Grebel has stated that these types of conductive polymer
 composites will have an advantage for some applications  since "conventional lead
 solder or other conductive metals, such as copper, possess mainly one degree of
 conductivity which is hard to vary under normal operating conditions."  One may
 postulate that conductive polymers may be made into shields for electromagnetic
 radiation by varying the amount of conductive filler in  the matrix. This is because the
 transmission drops sharply at very high frequencies as can be seen from Figure 7
 The frequency at which the transmission drops depends on the amount and type of
 tiller.                                                                    /r

 COST/BENEFIT ANALYSIS

      A cost/benefit comparison  of materials used in conductive polymer composites
versus conventional solder is shown in Table 2.  Graphite  is the conductive material in
the composite rather than the more common silver flake filler
TABLE 2.
Comparison of Materials Costs of Graphite-Based Conductive Polvmer
Adhesive and Tin/Lead Solder
          Material
                Cost/lb
Cost/g ftl
                                                      Density (a/ccl
                                                        Cost/
        Graphite fibers
                               17.50
                               0.039
                                                            .0
                                                           0.078
 EJ.     Paniculate graphite
 	powder	
                   1.30
  0.0029
                                              2.25
                             0.007
       Conductive graphite
       fiber/powder
       composite (45% fiber)
                  10.00
  0.022
                                              1.6
                             0.035
 P.    Tin/Lead (63%/37%)
                  90.00
                                           0.020
                                              8.17
                                                                       0.16
       Savings per cc: (D-C)
                                      102

-------
      Assuming that a typical circuit board requires approximately 6 cubic centimeters
of solder to complete all connections, and that a typical production run would consist
of about 1,000 boards, the cost savings by the switch from tin/lead to a graphite
powder-filled conductive polymer composite would be about $150, as indicated in line
E  The savings from the switch to the higher conductivity graphite filament would be
about $125, also shown in line E.  For a more effective composite, the powder and
fibers should be combined for optimum packing.  This should provide a materials cost
savings for the 1,000 boards of between $125 and $150.  While this figure may not
represent a large savings for 1,000 boards, significant savings may be realized over
large production runs or when a larger volume of  solder per board is required.
Environmental benefits include the elimination of lead and heavy metals from the
process; elimination of solvents used in conventional solder methods; and freedom
from flux residue on the boards that would require solvents plus labor costs to clean.


CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

       Polymer based solder material is a practical substitute for metal-based solders.
Additional development work will be necessary to define all parameters for achieving
high electrical conductivity and defining appropriate manufacturing and application
methods.  Tests indicate that other metal  replacement opportunities exist for such
applications as antennas and electromagnetic shields.

       Large or small organizations that are producing or fabricating electronic parts
may be able to replace their metal-based  solder material with filled polymer systems,
depending on the resistance requirements of the  applications.

Barriers

       The polymer solder material may not be as universally accepted as the
traditional tin/lead-based solder, mainly because  the higher conductivity of metals
(resistance 0.4 to 0.6 ohm) in comparison with the conductive polymer composite
coatings and adhesives (resistance  5 to 10 ohm). Also, at this stage of development,
the fillers are not dispersed as uniformly as will be desirable for many future
applications.

Potential Solutions

       Much higher conductivity of these  polymer-based solder materials can be
achieved by additional selection and optimization of filler ratios, but the present
product can probably be used in capacitors and  high voltage applications where very
high conductivity is not required.  Further investigations with improved dispersion
 methods, such as twin screw extruders and Banbury mixers, should be conducted.
                                       103

-------
                    COMPOUND ADIABATIC AIR CONDITIONING
                              FOR TRANSIT BUSES

                                       by

                                Jamends F. Mattil
                              Climatran Corporation
                              Englewood, CO 80155
 ABSTRACT
       Adiabatic air conditioning (AAC) is a water-based cooling process that requires
 minimal energy and uses no refrigerants, such as chlorofluorocarbons (CFCs)
 Compound adiabatic air conditioning (CAAC) is an advanced, two-stage version of the
 same general process, using indirect-direct cooling technology.  In the indirect staqe
 air is cooled without increasing its moisture content.  During the direct stage the air is
 cooled further by evaporating moisture into the air stream. The operable chemical  is
 each phase is water.

       This project assessed the performance of a CAAC system installed on a bus
 and operated under a variety of climate conditions representative of summer desiqn
 conditions in selected metropolitan areas.                                     y


 INTRODUCTION

 PROJECT DESCRIPTION

       Air cooling is a physical process governed by the laws of thermodynamics
 The relationships between air temperature and water content are well known and are
 described on psychrometric charts, which  provide a foundation for understandinq air
 conditioning performance.                                                y

       In a traditional refrigerant air conditioning system, a refrigerant gas is circulated
 in a closed system.  The gas is pressurized and liquified by a compressor   When the
 pressure is released, the liquified refrigerant quickly evaporates, absorbing heat from
 its surroundings (an evaporator coil).  An air stream is passed through an evaporator
 coil and the air is cooled sensibly,  meaning that no moisture is added  The  absorbed
 heat is carried by the gas until it can be rejected in a second process  (a  condenser
 coil).  A typical system contains two independent sets of blowers, motors and coils
 associated components, and a compressor.

 rr.iv ,H I*^£diab1tic' or' evaPpratlve> air conditioning (AAC) unit, air and water are
 mixed, and the water evaporates, absorbing heat from its surroundings.  This process
 occurs naturally, and pressunzation is not required; the energy requirements are thus
 much tower than those of a refrigerant unit.  The amount of cooling accomplishld
depends on the moisture content of the incoming air.  Dry air can hold more water so
more cooling can occur. Humid air already contains some moisture and  thus has less
evaporative potential. Evaporative cooling has been limited to dry climates for this
reason.
                                     104

-------
      A compound adiabatic air conditioning (CMC) system treats the supply
(incoming) air in two stages. In the first stage, the supply air is cooled sensibly. In
this project, a crossflow, wet-surface/dry-surface heat exchanger was used (although
other methods can be employed). On the wet side, air and water are mixed, the
water evaporates, the air temperature is reduced, and the humidified air is exhausted
to the atmosphere.  Supply air is passed through the dry side, its  heat is absorbed
through the heat exchanger walls, and its moisture content is unchanged.

      In this process, the dry bulb and wet bulb temperature of the supply air are
both reduced, so that the leaving air has an entirely new set of conditions that
increase the overall evaporative potential.

      The cooling capacity of any air conditioning system is expressed by the
following formula:

            BTUh  =  (CFM) (1.08 x elevation factor) (delta T)

where:
•     BTUh = cooling capacity, or load
•     CFM = air volume in cubic feet per minute
•     elevation factor = 1.00 at sea level (lower at altitude)
      delta T  = difference between  the temperature of the air entering and leaving the
      subject space

      For example, on a bus where the entering air is 60°F and the return (exhaust)
air is 80°F, the delta T is 20°F.  If the air conditioning unit delivers  3,000 CFM, the
cooling capacity is 64,800 BTUh.

      Transit buses present an extremely challenging cooling situation and require
very large, energy-intensive refrigerant air conditioning systems.  This is due  to high
occupancy, large glass surfaces, engine heat, and extreme infiltration from opening
and closing doors.  Low energy CMC systems can provide equivalent cooling along
with greater air volume and air movement, both desirable features.  The technical
challenge is to maintain acceptable performance under moderate and high humidity
conditions.

      The two-stage cooling process used in the CMC system can supply  air to a
given space that is both cooler and  drier than what is possible with a single  stage
MC system.  This capability can make it possible to maintain comfort in higher
humidity conditions and could substantially increase the geographic range where this
low energy, CFC-free technology can be successfully employed.

Unique Process Features/Advantages

      The Climatran CMC system  has two main elements: the actual cooling unit that
mounts on the rear of the bus, and  a water tank. The main cooling unit is located in
the upper rear area of the bus where one might expect to see a rear window.  This
unit consists of two independent sets of blowers and  motors, a water circulation pump
and distribution system, an electrical control  panel,  rigid evaporative media, and a
crossflow plate heat exchanger. The water tank can be mounted on the bus
                                       105

-------
  uriderbody near the front axle; the tank supplies water to the cooling unit with a
  transfer pump that is activated by a liquid-level switch in the cooling unit.

  th^ ™Ihe u,ni?ue comP°n?nt.isu the crossflow, wet-surface/dry-surface heat exchanger
  that pre-cools the supply air without increasing the moisture content. This is called
  sensible" cooling. As the dry bulb temperature of the air is reduced, the wet-bulb
  temperature is also reduced proportionally.

       After the air leaves the heat exchanger, it is passed through an adiabatic
  cooling stage where water and air are mixed and evaporation occurs  The  air
  temperature is reduced in proportion to the amount of water evaporated  If the
  LnHiohI?9 a'r Is drv; ™°re water can be evaporated, resulting in  greater cooling.  In an
  adiabatic system, the air temperature will be reduced until the air reaches saturation
  at which point the air temperature will equal the wet-bulb temperature of the entering
  air.                                                                        **

       The heat exchanger serves two functions.  As noted above, the crossflow heat
  exchanger reduces both the ambient wet-bulb temperature and the dry bulb
 fh?nPfhratureK- He-nce> trl? usecon.d stage' adiabatic cooler receives air that is  cooler
 pSential         air and has a lower wet-bulb condition that results in greater cooling


       The unique attributes of a CAAC system produces adequately cool air in
 c irnates otherwise unsuitable for standard evaporative air conditioning thereby
 eliminating the use of CFCs  and using little energy                   uiweoy
 APPLICATION

 Process Replaced
       Motor vehicle air conditioning systems have traditionally contained CFC-12
 refrigerant, commonly known under the trade names "freon" and "R-12 " This chemical
 has been one of the major causes of stratospheric ozone depletion and rorffibSS to
 global warming. EPA identifies CFC-12 as a class I ozone-depleting substance  CFC-
 12 can be operated in equipment using flexible, rubberized tubing.

 HPIFP ooThwHr^hf S 1,°me aPP|ications. such as transit buses, have begun using
 HCFC-22 (hydroch profluorocarbon , a somewhat different chemical compound but
 with properties similar to CFC-12.  Although less destructive than CFC 1 2  HCFC 22
 still releases chlorine molecules that interact with and destroy ozone molecules in the
 stratosphere; it is also a potent "greenhouse" gas.  EPA identifies ; HCFC-22 as fa dais
 II ozone depleting substance. This compound is very elusive and dfficult to contaS
                                                                         3  '
             C        nr"                      r PPn9  acansbject
leakage.  HCFC-22 has greater cooling capacity than CFC-12 and it too  is schedulPd

 00n phaSe'OUt under the Montreal Protoco1 ™<* the Clean Air Act
      Traditional air conditioning manufacturers have recently introduced a refrinprant
designated as HFC-134a (hydrofluorocarbon) that contains no cS Srine* atom? and
may not destroy ozone molecules.  HFC-134a has less cooling capacrty than efther
                                     106

-------
CFC-12 or HCFC-22 and is more difficult to contain and maintain, which may result in
excessive, unintentional discharges and  resultant consequences.

Wastes Prevented

      Although new regulations require  mechanics servicing air conditioners to use
recycling equipment, refrigerants still may leak during normal unit operation.  Leakage
is the single most common reason that equipment is serviced. The unintentional
release of ozone-depleting chemicals is  likely to continue, despite the use of recycling
equipment.

       Refrigerant air conditioning is also very energy intensive. A typical bus air
conditioning system operates in a range between 15 to 30 horsepower; 21
horsepower is considered as the standard value. A study by the Federal Transit
Administration found that refrigerant air conditioning results in an increase in fuel
consumption of about one gallon per hour of  operation.  Depending on the operating
area, this can amount from 600 to 4,500 gallons per bus per year.

       Since adiabatic systems operate on about 85 percent less power than
refrigerant equipment, substantial fuel savings are possible, which would reduce
carbon dioxide emissions as well as  other combustion by-products and particulates.

Cross  Segment Uses

       CAAC systems are already available  for residential, commercial,  and industrial
applications.  This technology could  be  applied to trucks, cars, and other
transportation vehicles.


PROCEDURE

DEMONSTRATION

       After the CAAC system was installed on the test bus, the bus was driven  to a
variety of locations to conduct performance tests under different climate conditions.
The 1992 summer weather was  unusually mild in many parts of the U.S., which
complicated route selection.  The main goal was to conduct some tests at dry bulb
temperatures around or above 90°F with coincident wet bulb temperatures of about
72°F or higher. Such tests would be representative of typical design conditions in
most of the northcentral, northeastern, and  middle Atlantic states.

       The original plan to test the bus in cities such as Chicago, Detroit, Pittsburgh,
New York, and Philadelphia became impractical due to the mild weather. Instead, on
the first trip, the bus was driven from Denver eastward to Salinas, Kansas, then
southward through Witchita and Oklahoma  City.  The return trip went west through
Abilene, north to Lamar, and back to Denver,  a total of 1,671  miles. The itinerary for
the second trip also began in Denver, proceeded south to Albuquerque and El Paso,
and west to Tucson and Phoenix where the demonstration was completed after 1,498
miles.
                                      107

-------
       Outside temperature and humidity measurements were taken at irreqular
       JsJ^d| when,suitablef a complete cooling performance test was conducted  A
 total of thirteen cooling tests were conducted. During each of these tests the bus
 would remain stationary, reducing the number of variables.

       The vehicle was equipped with thermocouples at five fixed points:

       Probe 1 measured the ambient dry bulb temperature
       Probe 2 measured the supply air temperature leaving the CMC system
       Probe 3 recorded the interipr space condition in the rear of the coach
       Probe 4 recorded the condition in the middle of the coach
       Probe 5 recorded the condition at the driver's seat

       The probe  locations were identical to those in prior tests (separate from this
 project)  of an AAC system.  Figure 1 shows the top and side view of the bus with
 temperature probe locations.

 F:igure 1.    Top  and Side View of Test Bus with Temperature Probe Locations
                /ENTRY
                ^-DOOfl
                                               _Q_
                           SEATING
              ®
SEATMO
                                              AIRFLOW
               DRIVER
                          PLAN VIEW WITH TEMP. PROBE LOCATIONS
                          S06 VIEW WITH TEMP. PROBE LOCATIONS
      After recording conditions with the air conditioning system off the tests were
begun by activating the CAAC system and recording temperatures at the 5 probe
locations at 2 minute intervals while the bus remained stationary  The tests were
                terior conditions stabilized. Temperatures were recorded on a multi-
                              bU'b temPeratures were measured with a Pyschrodyne
EVALUATION PARAMETERS

      There is a difference between refrigeration and comfort air conditionincr the
   Pr^nfnV11^^^ toumaintajn low temperatures to prevent spoilage  and the
latter is intended to maintain human comfort.  Either may be used in process
applications or to maintain specialized environments.
                                      108

-------
Comfort

      Vehicular air conditioning is intended to maintain human comfort for
passengers. However, the notion of "comfort" is elusive and subject to many
variables.  Four basic elements affect human comfort: temperature, humidity, air
velocity, and air volume.  Other factors include clothing, activity level, length of
exposure, individual transpiration rate, and individual perceptions.

      The American Society of Heating, Refrigeration, and Air Conditioning Engineers
(ASHRAE) has developed conditions that generally describe the human comfort zone
in terms of temperature and humidity.  In his book, The Evaporative Air Conditioning
Handbook, Dr. John R. Watt published a revised comfort zone that also considers air
velocity.  University studies have addressed human perceptions of comfort.
Unfortunately, in practice, little regard is  given to such quantifications of comfort;
instead the operable policy appears to be the colder, the better.

      According to studies conducted for ASHRAE, most human  comfort occurs
between 70° and 80°F.  Fewer people are comfortable at either extreme of this range
and the greatest number of people experience comfort between 74° and 76°F.
Additional studies by Kansas State University (KSU) have found that if humidity levels
are high, lower temperatures are needed to maintain comfort, and if humidity is low,
people can remain comfortable at higher temperatures. If air velocity or air volume is
low, lower temperatures are needed to maintain comfort, and if they are high, comfort
can be maintained at higher temperatures.  The KSU studies also  found that human
tolerances to extreme conditions are greater when exposure is for shorter time
periods, and that perceived discomfort increases over time.

      In this project, the cooling performance of the CAAC system was measured
against both the ASHRAE and the Watt  comfort zones.  The ASHRAE  Comfort Zone
indicates that human comfort occurs between 70° and  80°F with 20 to 60 percent
humidity.  The Evaporative Cooling Comfort Zone, described by Dr.  John R. Watt, also
considers the cooling effect of air velocity and generally indicates that  human comfort
occurs between 70° and 83°F with 20 to 80 percent relative humidity with appropriate
air velocities.  The humidity aspect presents some distinct problems. In any type of air
conditioning system, the humidity will vary significantly throughout the  space, making it
almost impossible to address quantitatively.  Therefore, it will be addressed
qualitatively, which, admittedly, does little to convince skeptics.

Design Conditions

      Since climate conditions vary, equipment or application engineers must
consider these conditions when designing and sizing an air conditioning system.
ASHRAE developed and maintains data describing the historical design conditions for
many areas.

      Design conditions are  expressed  in terms of dry bulb and wet bulb
temperatures.  They are presented to give the design engineer flexibility in selecting
the appropriate configuration. In most cases, a system that can maintain comfort 100
percent of the time is considered inefficient and overly  expensive,  since extreme
conditions are relatively rare. ASHRAE design conditions are expressed in terms of
the percentage of ALL hours that occur  at or below the given design condition.


                                      109

-------
        In a laboratory, the maintenance of specific atmospheric conditions miqht be

    othP tim? ann^n^U^'^ Elght require that comfort be maintained almost
  nntn^-o T   °n lansit buses°r ?ther Passenger vehicles, comfort is desirable but
  not crucial to operations or productivity. Therefore, 2.5 percent dry bulb and 5
  percent wet bulb design conditions will be considered as standards.

  ASSESSMENT SUMMARY

       This project has shown that a CAAC system can  maintain a comfortable
  environment in moderately high humidity conditions.  However, insufficient data were
  accumulated to facilitate comprehensive documentation of cooling performance under
  a wide variety of conditions.  Since the test vehicle lacked adequate electhcal ^owlr to
  operate the CAAC system at its designed output, performance results are not
  JhTrioSIKe; ' A"lerviC2 Pf.rformance would have to be predicted from extrapolation of
  the results  Although this is certainly possible, this approach could be unlikely "to
  convince skept.cs that a CAAC system would be feasible in their particular lowrtton.

  COST/PAYBACK OF DEMONSTRATION

       .7h*? C°DSt °f tf?-e pri?ject was $57'077-13, with $25,000 provided by EPA through

                                     Business erant pr°9ram- and 
-------
each probe location.  The average Interior temperature is the average temperature
recorded for probes number 3, 4, and 5.  In evaluating temperatures at specific
locations, the key is probe number 5, which is at the driver's seat and furthest from
the incoming supply air.

      In all but one test, the CMC system maintained an average  interior temperature
within the traditional ASHRAE comfort zone.  In test Number 9, the  average interior
temperature (81.7°F) exceeded the upper limit of the ASHRAE comfort zone (80.0°F),
however the average temperature did fall  within the comfort parameters of the Watt
comfort zone.

      Test Number 9 was conducted at Billings, Oklahoma with an ambient dry bulb
temperature of 90°F and a coincident wet bulb temperature of 75°F. Measured air flow
was 2,400 cubic feet per minute.  The supply air temperature was 75.4°F, and the
maximum exhaust air temperature was 83.5°F.  Calculated cooling capacity was
21,000 EBTUh.  If air volume was increased to the design point of 4,000 cubic feet per
minute (by installing a larger alternator), the cooling capacity would increase to about
35,000 EBTUh.

      Since a 21,000 BTUh unit reduced  the average temperature  by 8.56°F, it could
be calculated that a 35,000 BTUh unit would reduce the average temperature by
14.3°F.  The exhaust temperature as measured on probe number 5 would be reduced
from 83.5°F to about 76.6°F. The CAAC system, if operating as designed, could easily
have maintained comfort under either comfort criteria.

      In referring to the formula for cooling capacity, one may see that the only way
to maintain the condition of a given space as the supply air temperature rises is to
increase the air volume.  Given sufficient air volume, comfort temperatures can be
maintained  at virtually any ambient dry bulb design condition in the U.S.

      Based on Climatran's analysis of the results of this project, a CAAC system can
maintain comfort at design conditions up  to, and including 90°F dry bulb temperature
with a coincident wet bulb temperature of 75°F.  Increasing the air volume of the
CAAC system of 4,000 cubic feet  per minute would increase cooling capacity
providing further assurance that comfort would be maintained at 75°F wet bulb
temperature and extend the dry bulb parameter significantly  above 90°F.  The CAAC
system could maintain interior  comfort conditions at 95°F dry bulb temperature, 75°F
wet bulb temperature, and possibly somewhat higher.

      The continental United States may  be divided into four general regions where
AAC or  CAAC technology could be successfully applied to transit operations, based
on the performance measured during this project.  Zone I, which includes most of the
western U.S., is characterized  by wet bulb temperatures less than 70°F.  Zone II
includes a narrow area in the midwest and most of the northcentral and northeastern
states.  This region has wet bulb temperatures of 70° to 73°F and moderate humidity.
CAAC systems could be appropriate, but lower-cost AAC equipment also could  be
used in  applications where occasional reduced cooling performance is tolerable.
Zone III, a high-humidity area in the southcentral U.S. that includes  Virginia, Kentucky,
North Carolina, Tennessee, Missouri,  and  parts of Texas,  is defined as areas having
wet bulb temperatures of 74° to 75°F.  A CAAC system would be feasible; AAC
systems are unlikely to be acceptable, except for short periods.  Zone IV covers the
                                     111

-------
  deep southeastern states where very high humidity results in wet bulb temperatures of
  75 F and higher.  The cooling performance results in this project were not evaluated
  under conditions representative of Zone IV weather, and thus did not prove that a
  CAAC system could maintain comfortable conditions at 75°F wet bulb temperature
  and higher.  However, comfort might be maintained at or above 75°F wet bulb
  temperature if the system air volume was higher than that tested in this project
  PRODUCT QUALITY VARIANCE
 *u    u "^ii? ?ualitv- considered as cooling capacity and efficiency, was maintained
 throughout the test program, although not at the performance level to which the
 equipment had been designed. The equipment suffered no mechanical failures  and
 cooling  performance was consistent relative to expectations under the variety of
 2!T^e^                    ™e supply air volume was significantly less than
 HoSfn; Ahh°U£Lh Fe. S.UfDplied air temPeratures were acceptable, the volume of air
 delivered to the vehicle interior was much lowef than planned
 PAAP   n™~    caPacitv .is ? function of both air temperature and air volume, the
 CAAC cooling capacity was significantly lower than designed.  Therefore  interior
 temperatures during testing were higher than they would have been if the air volume
 nfln npon ac riaei/-in£iH                                                 TVIWHUV*
been as designed.
 at *h  H*   each !tatlonarv C0olin9 test, the bus was put in motion.  The temperature
    hdVer  Seat ba™ mm comfortable, and air movement became more
 nn inoh                                 ,
 noticeable.  After installing a supplemental thermometer, the driver was able to monitor
  he temperature at the driver's seat while in motion, which confirmed that the
 temperature decreased further when the bus was in motion.

       '!? terst Tmbf rJ 1 (Tucs°n. Arizona), the driver's area reached a temperature of

                                    the bus beaan to move' the
 CONDITIONS THAT IMPACT PERFORMANCE

 m,-« * Thu CAAC aJu volume was measured and calculated at 2,400 cubic feet per
 SS± J±^r^tSlUIK^^ *S!!* * tuPP'y4;000 'ubic fef Per minute.
                                          at a level that would supply the desired
      The power supply from the existing bus alternator was 145 amps while the
power demand from the blower motors was 255 amps. The actual Kravlilabl.
was only about 60 percent of that necessary to operate the system Ksfgned

des'  Td6 a'r V0lufme was measured at 2,400 cubic feet per minute compared to the

shortage of electrical'power.
             ?*   if ontwas.identified, a new, higher capacity alternator was ordered
      tmoV' t  alter'?lftor,dld not arrive before the weather had deteriorated to
conduced       SS'     alternator was no* replaced, as no further testing could be
                                     112

-------
      While the bus was in motion, the location of the test equipment prevented the
driver (the only person on the bus) from monitoring the various temperature probes,
with the exception of the supplemental thermometer measuring probe number 5.

      The CAAC unit included a roof top deflector, or hood, to keep rain from
entering the unit. This hood had an air inlet opening that faced the front of the bus.
This enhanced the performance of the blowers and offset the shortage of air volume
(that resulted from the under-sized alternator).

      Future testing should document the effects of the air inlet and evaluate cooling
performance of the CAAC system when operating at its designed output. Such
increases in air volume would substantially enhance cooling performance and could
result in the ability to maintain comfort at and above 75°F wet bulb temperature.

TABULATION OF DATA

      The results of the 13 individual cooling tests are summarized in Table 1 and
shows the outdoor conditions during the tests and the cooling results in terms of the
average interior temperature.  To give perspective to the ambient climate conditions,
the last column indicates cities having summer design temperatures similar to the
conditions experienced during the test.

TABLE 1.    Summary of Cooling Performance Test  Results
Test Number
1
2
3
4
5
6
7
3
9
10
11
12
13
Test Type
AAC
CAAC
CAAC
CAAC
AAC
CAAC
CAAC
CAAC
CAAC
CAAC
AAC
CAAC
CAAC
Ambient
dbt/wbt*
88/61
88/61
91/67
88/74
89/67
90/67
84/69
85/66
90/75
90/73
93/62
96/61
104/62
Ave. Interior
Temp. (°F)
77.30
69.90
75.30
79.60
80.10
76.10
75.90
75.30
81.70
79.10
77.00
75.00
73.40
Location
Denver
Reno
Pasadena
New York
Burbank
Amarillo
Buffalo
Portland, OR
Atlanta
Philadelphia
Salt Lake
Albuquerque
Tucson
 ^obt/wbt - dry bulb temperature/wet bulb temperature
                                     113

-------
 COST/BENEFIT ANALYSIS

       In the report, "Evaporative Coolers for Transit Buses," published by the Federal
 Transit Administration (formerly UMTA), the University of Denver, Denver Research
 Institute conducted a life cycle cost comparison between refrigerant and evaporative
 air conditioning equipment.

       This report concluded that substantial fuel savings and maintenance  cost
 savings could be realized with evaporative equipment.  Fuel savings would  amount to
 approximately 1 gallon per hour of system operation.  Maintenance costs would be
 about $200 per year compared to $1,200 to $1,700 per year for a refrigerant-equipped
 bus. Although the reliability  of an  evaporative unit was found to be substantially
 greater than a refrigerant unit, no additional  cost savings were attributed to  this factor.

       A vehicle operator could expect maintenance cost savings of about $1 250 per
 bus per year with an evaporative system. Since this report was published  EPA has
 required CFC recycling and  prohibits known releases of CFCs  This would  likely
 increase maintenance costs  for CFC-based equipment above the levels noted in the
 FTA report.

       Operating hours range from 600 to 4,500 hours per year depending on
 geographic regions and the  type of service provided.  Assuming 1  800 annual
 ?^r?tH?g hours over a 5 month Period. an operator could save 1,800  gallons of fuel
 (at $1 00 per gallon) per bus, and  about $1,500 on maintenance, for a combined
 annual savings  of $2,300 per bus equipped with an adiabatic system.

       On new buses, the purchase price of  CFC systems and AAC systems are
 virtually equal, however adiabatic equipment could generate immediate savings due to
 lower fuel and maintenance costs.  The purchase price of CAAC units  would be
 approximately $1,500 more than either the CFC or AAC units; given monthly savings
 of about $500, the simple payback period for a CAAC system would be about 3
 operating months.

      Metropolitan transit operators receive federal subsidies for equipment
 purchases and operating shortfalls. These subsidies are traditionally 80 percent of
 cost  The operator's share of the CAAC premium is only 20 percent, or about $300
 which would payback in less  than one month. Both the transit operator and the    '
 federal government could save money by adopting  the AAC and CAAC systems.

      Existing buses generally operate CFC-12 air conditioning equipment and
 experience substantial leakage and major maintenance expenses, much of which is
 equipment replacement.  Retrofitting existing  vehicles could immediately reduce CFC
 emissions.  A bus could be converted to a non-CFC system for about $9 000
 (subsidized by $7,200 in federal funds).  Annual operating cost savings would be
 about $2,300 or more, since CFC equipment maintenance costs are higher as
 equipment  ages, and retrofits would be done to older vehicles.

      The local  operator could recover its $1,800 investment in about 4 months while
the entire retrofit cost could be recovered in about 20 operating months.
                                     114

-------
CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      This project showed the potential of CMC systems to provide comfort air
conditioning for transit buses under climate conditions that would have been
impractical for traditional evaporative cooling equipment.

      The CAAC equipment produced for and tested under this program functioned
with no mechanical or electrical failures, but did not perform as designed due to
shortage of electrical power on the test vehicle. Despite this significant deficiency, the
CAAC system was able to maintain  interior comfort conditions  as defined by one  or
both sets of criteria designated for use in this evaluation.

      Cooling performance data were collected during  13 individual cooling tests
conducted under a variety of climate conditions.  Testing showed that the cooling
capacity  of the CAAC system is inversely related to outside humidity (wet bulb
temperatures). Based on the data collected, the  CAAC system has the capacity to
maintain  comfort conditions at ambient conditions of at  least 90°F/73 percent humidity.
Since cooling performance suffers little from increasing dry bulb temperatures, comfort
can likely be maintained up to 95°F/73 percent humidity.

      Although cooling performance was affected by a shortage of electrical power
from the  bus alternator, one may calculate cooling capacity that could have resulted
using a properly sized alternator and extrapolate  expected performance  based on
actual test results.  In this case, the  CAAC system could easily have maintained
comfort at the above-mentioned  conditions.

      Based on the demonstrated cooling capacity of the CAAC system, this
technology could be employed in areas where design conditions are appropriate,
which would include all of the continental U.S., except for the general region east of
100 degrees west longitude and south  of the 36th parallel, with other localized
exceptions.

      Adoption of this technology could eliminate the use of ozone-depleting CFCs
from bus applications and would conserve fuel.

Barriers

      Barriers to the advancement and acceptance of this technology are  primarily
political in nature, rather than technical.  Technical improvements and cost  reductions
could be expected to coincide with market development, however AAC and CAAC
systems already offer substantially lower life cycle costs than CFC-based refrigerant
systems.   Opportunities exist for the federal government to adopt this technology and
to encourage its use by federally subsidized metropolitan transit operators.

      The cost of a CAAC system is currently about 25 percent higher than that of a
CFC system. CAAC technology  has no scale economics.
                                     115

-------
  /CTAX 7ransit.operators receive subsidies from the Federal Transit Administration
  (F FA) for equipment purchases and operation shortfalls.  This policy removes anv
  economic incentive to purchase cost-effective equipment or to operate more efficiently
  CFG equipment becomes more affordable and is often used in areas where the need
  for air conditioning is marginal.

       FTA refuses any assistance to AAC technology and has imposed testing
  requirements that block the use of this technology on new buses While the
  Department of Defense has approved the use of AAC technology, the General
  Services Administration will not specify the inclusion of such equipment on new
  vehicles, citing an executive order by President Reagan that minimized unnecessary
  requirements on private businesses supplying the federal government.

       The crossflow heat exchanger currently in use is the  only model available  The
  efficiency of this device could be higher, but no manufacturer produces an improved
 version, since there is virtually no market for it.  A higher-efficiency heat exchanger
 would result in lower supply air temperatures and lower humidity levels both of which
 would enhance performance and possibly reduce equipment size and cost.
                                 eXChan9er JS SUCh that * cannot
 in addd
           f~mPOTertS ^ *" ^^ is comP|icated and
 iow«io  ^°th MS T u CA^-C svstems C001 by increasing the humidity above ambient
 levels.  In very high humidity conditions, this adversely impacts interior comfort and
 requires very high  air volume levels to maintain acceptable temperatures.

 Potential Solutions
                   Purchase Pricue for CAAC equipment is currently about 25 percent
«          f°r CFC ®gu|Pment, the technology has no market share, no economies of
scale and no competitive pressure.  As the market for this type of equipment is
developed  the first cost premium could be eliminated quickly through natural
c^r*^^ n^^rY^ i^* TO^^T^^KO                                             ^^
economic factors.
          present subsidy program may discourage transit operators from pursinq
    -saving measures and encourages the use of CFC equipment. Through
wh»rL tht n°^f' C C Sys^s .become more affordable and are often used in areas
where the need for air conditioning is marginal.

      One approach to this dilemma is to amend the subsidy formula to remove anv
SWSwS,^^886^ C^-C an,d HCFC ^iP-^nt, whileyretainingTundTng ?of V
rn
-------
                      REDUCING HEAVY METAL CONTENT
                           IN OFFSET PRINTING INKS

                                      by

                                Roger Telschow
                                    Ecoprint
                            Silver Spring, MD 20910
ABSTRACT
      Ecoprint created a commercial offset printing ink using pigments with no heavy
metals, thus preventing pollution in three key areas: (1) in the waste ink produced by
a printer that must be handled as a hazardous waste;  (2) in the printed materials that
are landfilled or incinerated; and (3) in the sludge that  is created during the de-inking
and repuljping of waste paper fibers as they are made  into recycled paper. The result
of the testing throughout the project was the creation of a "palette" of colored inks with
a low heavy metal content.


INTRODUCTION

PROJECT DESCRIPTION

      Printing is an $80 billion a year industry in the United States and uses large
amounts of ink.  Colored inks contain many heavy metal-based pigments that are
applied to a myriad of printed materials that ultimately end up in the nation's waste
stream.  This project had five key components:

•     Tested the 11 primary colors of printing inks to determine the concentration of
      12 key metals

      Identified alternative pigments that were not based on heavy metal compounds

•     Formulated new offset printing inks based on these non-heavy metal pigments

•     Tested these inks in actual commercial printing conditions

•     Created a "palette" of non-heavy metal-based inks in primary mixing  colors

Unique Product Features/Advantages

      These non-heavy metal-based offset printing inks can be successfully used in
most commercial printing applications including the printing of brochures, newsletters,
direct mail promotions, letterhead, and many other products.  Ecoprint knows of no
other sheet-fed lithographic inks on the market that have tested as low for metals
content.
                                     117

-------
 APPLICATION

 Products Replaced
       The result of this project is the creation of inks with greatly lowered heavy metal
 content, thus preventing pollution in three key areas: (1) in the waste ink produced by
 a printing company that must be handled as a hazardous waste; (2) in the printed
 materials that are later landfilled or  incinerated; and (3) in the sludge that is created
 during the de-inking and re-pulping of waste paper fibers on their way to beinq made
 into recycled paper.
 Waste Prevented

       While Ecoprint cannot foresee direct cost savings to the printer (waste ink of all
 kinds still contains oily compounds, and as such will have to be disposed of as
 hazardous waste), but there will be savings on a more general, societal level if the  use
 of these new inks becomes commonplace. These savings will come in the form of
 significantly reduced pollution of air, water, and land. Incineration of waste papers
 printed with these new inks will produce ash and air emissions that are lower in
 copper and barium.

 Cross  Segment Uses

       Other sectors of the printing industry also may be able to use these inks
 although they would need to be reformulated to be compatible with the different
 machinery used. Additional applications include flexographic printing, engraving
 silkscreening, and other types of offset printing such as web printing (cold and
 heatset). A wide variety of products - from packaging materials to magazines - are
 printed with these different printing applications.

 PROCEDURE

 DEMONSTRATION

      Ecoprint first selected the metals to be tested by surveying a number of expert
sources - EPA's 33/50 Program, The Chesapeake  Bay Foundation, and the National
Toxics  Campaign Laboratory ~ for information on environmental toxins   The 12 taraet
metals  selected were:                                                      y

      Antimony
      Arsenic
      Barium
      Cadmium
      Chromium
      Copper
      Lead
      Mercury
      Nickel
     Selenium
     Silver
     Zinc
                                     118

-------
       Ecoprint then arranged for the testing of the 11 primary ink colors (neutral
 black, transparent white, reflex blue, process blue, green, yellow, warm red,
 rhodamine red, rubine red, purple, and violet) to determine which inks had
 problematic heavy metal contents.  Samples of the 11 inks were sent to the National
 Toxics Campaign Laboratory in Boston where the inks were tested for heavy metal
 content with a Perkin-Elmer Atomic Absorption 5100 spectrophotometer.  Certain
 samples were further tested for barium using Inductively Coupled Plasma Atomic
 Emission Spectroscopy (ICP), when false readings were generated by the presence of
 calcium.

       Ecoprint selected the target ceiling of 100 parts per million (ppm) for a heavy
 metal content goal. Tests revealed the problem ink colors to be reflex blue (over 200
 ppm copper), process blue (3,800 ppm copper), yellow (859 ppm zinc), green (3,300
 ppm copper), warm red (122 ppm barium), rubine red (150 ppm zinc),  and rhodamine
 red  (181 ppm copper).  The other colors tested below the 100 ppm level for the target
 metals.

       Ecoprint then experimented with alternative, non-heavy metal-based pigments,
 such as pigments based on calcium compounds or organic pigments.  These
 alternative pigments should match existing colors as closely as possible and should
 be compatible when mixed with the other components of the ink, such as the resins
 and oils.

       Since these reformulated inks are a proprietary product that will be
 subsequently marketed, the alternative pigments used cannot be specifically named.
 However, the pigments are sold by well-known pigment  manufacturers, but have not
 generally been used in printing ink formulations.

       The actual mixing of the inks was conducted by Ecoprint's subcontractor, Alden
 and Ott Inks. Newly formulated inks that contained the alternative pigments were then
 sampled and sent back to the laboratory to again test for metal content, again with the
 goal of formulating inks that tested below 100 ppm for each of the 12 metals.  Several
 colors of inks were reformulated in one trial; other ink colors required several trials to
 obtain the correct formulation. Only two colors, rubine red and rhodamine red, could
 not be reformulated, as pigments could not be found to  match the shade of these
 colors.

      Once achieving the desired test results for metal content, the inks were tested
 on a printing press to determine their printability.  As the inks were used in printing,
 their performance was monitored in several areas: drying time; absorption onto the
 paper surface; compatibility with printing plates, fountain solutions, cleaning agents,
 and solvents; and "holdout" on the sheet after printing to determine if density of color
 remained strong.  Inks were also mixed to see if a sufficient number of different colors
 could be created to satisfy most commercial requirements.

      Approximately 12 press tests were conducted on  a Komori Sprint, 1978 model,
2-color press.  The machine was retrofitted with an Epic  Delta Dampening System
which utilized a separate water form roller with an oscillating bridge roller that contacts
the first ink form roller.  Other specifics included:
                                      119

-------
       Running speeds:  Between 5,000 to 9,500 sheets per hour

 •     Blanket:  Reeves Vulcan 714 Compressible

       Plates: 3M Viking subtractive

       Fountain Solution:  Rosos G-C #1J One-Step concentrate, mixed as follows - 6
       ounces concentrate, 2 ounces gum arable, using water from a reverse-osmosis
       unit; solution recirculated through Royce refrigeration unit

 •     Room Temperature: Between 68 and 75 degrees F

       Print Run Length:  1,000 impressions up to 20,000 impressions

       Papers:  All uncoated and recycled, but varying in weight from 60 pounds offset
       to 80 pounds cover

       Ink Coverage:  From 5 percent to 50  percent coverage, including solids
       screens, and traps, as well as line art

       The press was  set up as for the "regular" inks, with no change in work habits.

 Cost of Demonstration

       EPA's  contribution to the project through the Pollution Prevention By and For
 Small Business Grant  Program was $25,000; Ecoprint's contribution was $6 749  for a
 total project cost of $31,749.

       While the  cost of laboratory analysis was higher than first estimated, it was a
 vital part of the project.  Printing compatibility was much more trouble-free than
 expected, thus lowering  costs in this category.  Ecoprint was pleased with the overall
 results of the project, considering its modest scope and the warning by several
 industry sources that the goals may be "impossible to achieve."


 RESULTS AND DISCUSSION

 PERFORMANCE RESULTS

 Press Test Results

      No difference was detected between the reformulated and the original
formulation inks in the following categories:

      Interaction with  plate surface:  No  problem  was encountered with plate
      sensitivity in non-image areas, plate blinding, or premature plate wear.

      Mixability of inks:  More study would be desirable to determine if pigment
      strength poses any problem, particularly with colors mixed with a high
      proportion of transparent white.
                                      120

-------
 •     Performance on press:  No stripping on ink rollers was seen; wash-up
       procedures and interaction with fountain solutions (running alcohol-free posed
       no problems) were normal; and no tendency of inks to "emulsify" was detected.

 •     Printing on paper:  Good absorption onto sheet surface; good trapping with
       other colors; drying time averaging the same as original formulation inks.
       Colors seem to be lightfast indoors (i.e., office lighting) for at least two months
       (and probably much longer, although long term tests were beyond the scope of
       this project).  No tests were conducted where ultraviolet light (i.e., sunlight) was
       exposed to the printed material for extended periods of time.

 •     Folding of printed sheets:  While occasional marking (where pullout rollers
       collect ink and redeposit it onto the sheet as a light mark) was noted, this
       problem was no worse than with the original formulation inks.

 TABULATION OF DATA

       The metal contents of the five ink colors that were reformulated are shown in
 Table  1.


 TABLE 1.    Metal Content in Inks before and after Reformulation (Parts per Million)

Metal
Antimony
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
Reflex Blue
Before
*<10
6.30
8.80
0.35
,2
205,0
1.50
<0.1
5.28
<10
<0.2
4.30
After
<2.5
<0.t
55.6f
1,14
1,00
1.60
1.00
<.05
1.40
<0.1
<<05
10.1
Green
Before
<10.0
<10.0
76.00
0.86
8.60
3300.0
<0.1
5.10
4.66
<10.0
<0.4
59.00
After
42.00
2.00
<36.0
0.50
15.90
10.60
<2.0
22.00
6.60
<36.0
1.20
3.70
Warm Red
Before
7.00
17.00
122.00
0.52
3.80
1.22
<0.3
0.09
5.51
<10.0
<0.2
26.00
After
<5.0
<3.6
4.50
3,04
12.40
2.50
8.90
<0.2
1<96
<9.0
<3.6
63.20
Yellow
Before
* <10
4.20
11.40
<0.04
<0.4
2.39
<0.5
1.10
6.08
<10.0
<0.2
859.00
After
<2.0
12.00
<22.0
<0.1
<0.07
<0.07
3.60
0.73
<0.1
<22.0
0.40
9.00
Process Blue
Before
<10
<0,3
6.60
0.33
3,30
3800
0.70
<0.1
6.53
<10
<0.2
6.20
After
19.0
11.0
29.0
0.20
11.9
9.40
<1.0
20.0
5.30
<11
0.30
12.2
  •A muiuaiws uiai uuricenirauon is lower man me limns 01 aeieciion
f a second test indicated less than 6 ppm

      In reflex blue, copper was nearly eliminated, dropping from 205 ppm to less
than 2 ppm.  Barium did increase to 55 ppm in the first analytical test, but stayed at
less than 6 ppm in a second test.

      In reformulating yellow, the first analytical test resulted in a reading of 859 ppm
for zinc.  While the pigment was not changed, Ecoprint suspected that yellow was
                                       121

-------
being mixed in the presence of another contaminant.  A more careful mixing of the
yellow did result in the near elimination of the zinc to 9 ppm.

      The metals content of the inks that did not require reformulation (as all metals
were below 100  ppm) is shown in Table 2.
TABLE 2.    Metals Content of Inks not Requiring Reformulation
Metal
Antimony
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
Trans. White
15.00
17.00
* <0.5
0.99
2.10
1.09
2.10
<0.1
3.80
<10
<0.3
3.10
Neutral Black
<10
15.00
<0.5
0.67
1.10
4.59
<0.6
0.50
4.98
<10
<0.3
8.30
Purple
<10
<7
6.60
0.20
46.00
1.40
1.30
2.60
4.00
<10
<0.3
8.70
Violet
<10
4.10
12.00
0.43
2.30
79.90
3.50
<0.1
5.16
12.00
<0.2
5.10
  
-------
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
2.16
3.00
0.01
6.82
<10
<0.2
150.00
181.00
1.30
<0.04
5.78
<10
0.07
9.10
  
-------
       The cost breakdown of a sample print job is outlined in Table 5: an 8 page
 newsletter, 70 pound opaque recycled paper, 50,000 copies, prints black and one
 color ink, folded to mail.  (These figures are taken from cost breakdowns provided by
 the Printing Industries of America 1988 "Financial Ratio Study.")

       Considering all types of commercial print jobs, ink constitutes, on average, 1.62
 percent of the total price of the job.  A three-fold increase in ink price would increase
 the cost of the job by slightly less than 3.4 percent.  Small jobs would probably
 increase by a higher percentage as would jobs using a large amount of ink coverage.
 TABLE 5.    Cost Breakdown of Sample Print Job
Cost Factor
Total Sales Price
Total materials, including paper,
outsider services, but excluding ink
Factory payroll
Factory expenses
Administration, interest, and selling
expenses
Profit before taxes
INK
Price
$5,000
$1,782
$1,283
$664
$1,039
$151
$81
Percent of Total
Cost
100.00%
35.64%
25.66%
13.27%
20.78%
3.03%
1.62%
CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      The reformulated inks are now available for use in sheetfed offset printing, and
can easily be mixed in larger quantities by Alden & Ott. Since there are many
variables in printing and before making wide guarantees of press performance to
customers,  Ecoprint is conducting even more press tests.

      Any sheetfed commercial printing establishment should be able to use these
inks. Even  web printers (cold-set) should be able to test these inks with some minor
modifications in their formulation.
                                     124

-------
Barriers

      Not every color in the Pantone Matching System (PMS) color book can be
mixed, as Ecoprint does not yet have substitutes for two primary mixing colors, rubine
red (high in barium and zinc) and rhodamine red (high in copper). PMS is the most
widely accepted color matching system in the industry, and some customers will need
colors that cannot be mixed using non-heavy metal inks.

      One solution to the PMS matching dilemma is to continue the research to find
substitutes for rhodamine red and rubine red. Should this not be possible, customers
can choose from the PMS colors that can be mixed with the 9 colors that all passed
the 100 ppm metals test; this still provides a significant palette of colors with which to
work.

      The cost of some alternate pigments may always be slightly higher, but if these
inks were to be adopted by the industry,  the price of the pigment may drop due to
large-scale production. Print buyers may need to be educated to use less ink
coverage on the sheet to control ink costs.  For example, one simple way to
accomplish this is to discourage designs  that cover an area with 100 percent ink and
then "reverse out" a headline so that the letters appear in white. Lighter ink coverage
has the additional advantage of being "more environmentally sound," as less ink on a
sheet makes it more easily recycled.
                                     125

-------
                       REUSING ZINC PLATING CHEMICALS

                                       by

                                Douglas Brothers
                                Global Plating, Inc.
                               Fremont, CA  94538.
 ABSTRACT

       Global Plating, Inc., a full service plating and anodizing job shop, proposed to
 recover zinc chloride and other chemicals used in zinc plating through ultrafiltration
 and reverse osmosis membranes.  This would reduce the amount of hazardous waste
 generated, remove chlorides from the wastewater, and permit reuse of the reclaimed
 metals. Global Plating assessed the effectiveness of the membranes, but technical
 difficulties prohibited the actual recycling of zinc and plating bath chemicals.


 INTRODUCTION

 PROJECT DESCRIPTION

       Zinc plating of steel is an inexpensive method of protecting steel from
 corrosion.  Zinc plating, when combined with organic brighteners, provides an
 attractive, bright finish to the workpiece.

       In Global Plating's production line, zinc plating solution is dragged out of the
 zinc tanks on freshly plated parts. The parts are then rinsed with water in the rinse
 tanks that are overflowed at a rate of 0.5 gallons per minute to maintain a low zinc
 concentration flow rate and reduce water volume.  The overflowed rinse water is piped
 to a collection sump  where it is pumped  to a waste treatment system.

       In the treatment system, zinc and metals from other plating processes  are
 precipitated from the wastewater.  The precipitate is pumped to a filter press  where it
 is formed into cakes  1 inch by 24 inches by 24 inches and containing 30 to 40  percent
 solids. The cakes are placed in a sludge dryer where water is removed, leaving a
 powder containing 80 percent solids.  This powder is sent to a certified recycler for
 metals recovery. The discharged wastewater must meet local discharge requirements
 of 2.6 parts per million zinc.

      Global Plating  proposed to reclaim the zinc and organic brighteners on-site
through the use of reverse osmosis (RO) membranes.  This process would allow the
materials from the zinc rinse tanks to be returned to Global Plating's process  In
plating approximately 5,090 square feet of steel per day, Global  Plating uses about
9,000 gallons of water daily for rinsing the plated parts. Zinc levels  in the wastewater
averages 100 parts per million (approximately 0.75 pounds) of zinc  per day.
                                      126

-------
      In the RO system, water is pumped through membranes that allow only the
passage of water molecules. Other molecules are rejected and may be recovered. In
the zinc plating process, materials that will not pass through the RO membranes are:

      2[inc chloride
      Potassium chloride
      Etoric acid
      Zinc brightener
      Zinc brightener carrier
      Organic contaminants
      Inorganic contaminants

      The RO membranes can be plugged from contaminants or incompatible
chemicals.  The correct membrane must be used and contamination must be
controlled.  Even under ideal conditions, the membranes can become plugged and
must be cleaned periodically.  Prefiltering the solution to be treated can prevent
premature plugging of the membranes.  Global Plating installed a 5 micron filter
cartridge and ultrafiltration (UF) membranes in the process stream before the solution
entered the RO membranes. UF membranes  are similar to RO membranes in that
they have controlled size pores that allow certain sized particles to pass through. The
use of UF membranes extends the life and period between cleanings of the RO
membranes.  With the addition of the extra filtration, the ratio of zinc concentrate to
elutant remained at acceptable levels of above 8:1 for two months after installation of
the additional membranes.  After two months,  the ratio rapidly dropped to 4:1,
indicating that the membranes must be cleaned  or changed.

      Zinc rinse water was passed through the  RO membranes that only allow water
to pass through. The filtered water was  returned to the zinc rinse tanks at 2 gallons
per minute. The concentrate was recycled at  16 to 20  gallons per minute in the RO
process until the conductivity of the concentrate reached a preset level. At this time, a
solenoid valve opened and the concentrate was returned to  the zinc plating tank at 2
gallons per minute.

Unique Product Features/Advantages

      The RO membranes can recover zinc and the organic brighteners more
efficiently than precipitation.  The process also can be conducted  on-site so that the
zinc and brighteners may be returned to the plating process.

Process Schematic

      Global  Plating's concentrated zinc recovery system is shown in Figure 1.
                                     127

-------
Figure 1.    Concentrated Zinc Recovery System
                        RINSE TANK
                        RECLAIMED
                        WATER f 2
- CASCADE ->
            RINSE TANK
            DRAG OUT
               #1
 ZINC
 RINSE
WATER
                 RECLAIMED
                RINSE WATER
         RO UNIT
               WASTE ZINC
              CONCENTRATE
              TO CLARIFIER
                                      ZINC TANK
                                         #1
                       I
                    ZINC TANK
                       *2
                                            EVAPORATOR
                       ZINC STORAGE
                       CONCENTRATE
                ULTRA
              FILTRATION
                                     128

-------
APPLICATION

Process Replaced

      The RO process may be used to replace or reduce the size of a conventional
precipitating, clarifying waste treatment system. The conventional clarifying,
precipitation system is shown in Figure 2. In Figures 1 and 2, the illustrated process
takes concentrated zinc metal from the drag-out tank and adds it back to the zinc
bath. This method reduces the amount of zinc in the waste treatment unit, therefore
reducing the amount of metal that must be treated and disposed.


Figure 2.    Conventional Clarifying, Precipitating Waste Treatment and Zinc Process
            Line
                         4.5*
                     76.5'
                                  CLEANER
                                  CLEANER
                                   RINSE
                                    ACID
                                    ACID
                                   RINSE
                                ZINC SOLUTION
                                ZINC SOLUTION
                                ZINC SOLUTION
                                  RINSE *1
                                  RINSE *2
                              BLACK CHROMATE
                                   RINSE
                              CLEAR CHROMATE
                              YELLOW CHROMATE
                                   RINSE
                                  HOT RINSE
                                    -12'-
                                                  AIR BLOWER     TO ROOF
                                                    t
               t
              ATMOSPHERIC
              EVAPORATOR
           CONCENTRATED
            ZINC PLATING
              SOLUTION
                                               ZINC RINSE WATER
t
             R.O.
           SYSTEM
    R.O. ELUTANT
    (PURE WATER)
                                      129

-------
The conventional waste treatment process is illustrated in Figure 3.  Zinc hydroxide is
processed through a filter press and a dryer before being disposed.  The wet volume
of the zinc hydroxide can be reduced 4 to 5 times with this process, which results in
lower freight and disposal charges as less volume (i.e., less water) would be shipped
arid disposed.


Figure 3.     Waste Treatment (Clarifying Precipitation) Process
                                                               TO SEWER
                                           PH
                                       CONTROLLER
                FROM ZINC
                 RINSE
                         FLOCCULANT  SODIUM
                                  HYDROXIDE
                           Or
                                                                    FILTER
                                                                    PRESS
                                      SLUDGE
                                       DRYER
                       SHIP TO
                      RECYCLE
MOVE PRESS CAKE
TO SLUDGE DRYER
HOPPER WHEN FULL
                                       130

-------
      The RO zinc recovery process is completely separated from the clarifying,
precipitating waste treatment system, the latter being common in metal plating
operations. When Global Plating uses the precipitation system, the resulting sludge is
sent for metals recovery rather than disposed as hazardous.

Cross Segment Uses

      The RO process is applicable to a variety of plating operations, such as for
bright nickel, where metals recovery is desired.


PROCEDURE

DEMONSTRATION

      Following installation of the RO system as shown in Figure 1, trials were
conducted during normal production on the zinc plating line.  The plating process was
not changed. The only alterations to the plating line were the piping to the RO system
and to the atmospheric evaporator.

      The atmospheric evaporator was located  beside the zinc plating tank.  Zinc
plating solution was pumped to the  evaporator where it passed through  a series of
large baffles.  Air was blown down through the baffles, evaporating the water.  The
evaporation rate depended on:

•     Air volume
•     Air temperature
•     Zinc plating solution temperature
•     Zinc plating solution volume
•     Baffle surface area

      The atmospheric evaporator must evaporate enough zinc plating solution to
accommodate the RO concentrate returning from the  RO system. Without sufficient
evaporation, the RO concentrate automatically diverts to the UF collection sump tank.
When this sump is full, the RO concentrate can be diverted to the pretreatment
storage tank. When these are full, the RO concentrate is automatically overflowed to
the clarifying, precipitating waste treatment system where the  zinc is precipitated into
sludge.

      The RO system operates as follows:

1.     City water enters a water softener to remove any water hardness from the
      make-up water to the RO system (i.e., all  make-up water should be de-ionized
      or RO water).

2.     Zinc rinse water from the #1  zinc rinse tank flows to the UF collection sump
      tank at 2 gallons per minute.

3.     This water is pumped by a 0.5 horsepower pump through the 5 micron filter
      cartridge, through the UF membrane, and collected in the  pretreat storage tank.
                                      131

-------
 4.    The water is pumped by a 5 horsepower pump at 20 gallons per minute at 200
       pounds per square inch through the RO membranes.

 5.    Water molecules pass through the membrane and return at 2 gallons per
       minute to the #2 zinc rinse tank. Because RO membranes are not completely
       efficient, approximately 20 percent of the chemicals leak through the
       membranes and return to the #2 zinc rinse tank.

 6.    The rejected chemicals from the RO membranes are directed back to the
       pretreat storage tank for another pass through the RO membranes. This
       sequence is followed until solution conductivity levels trigger the opening of a
       solenoid valve.  The concentrated reject solution is sent to the zinc plating tank
       at 1 gallon per minute.  The conductivity controller is set at 2,000 parts per
       million. The solution returning to the zinc plating bath is concentrated about 4
       to 6 times.

 Zinc Plating Tank Adjustments

       Global Plating uses an acid chloride zinc solution with the following
 characteristics:

       Zinc chloride: 4 to 6 ounces per gallon
       Potassium chloride:  17 to 21 ounces per gallon
       Boric acid:  3.5 to 5 ounces per gallon
       Brightener:  0.05 to 0.1 percent by volume
       Carrier:  3 to 4 ounces by volume
       pH:  4.5 to 5.5
       Temperature:  65 to 90°F

       This type of solution has been standard for approximately 20 years and has
 been a replacement for zinc cyanide solution.

       The  acid chloride zinc solution must be run with organic additives that modify
 the crystalline structure of the plated deposit to improve its quality. The brightener
 can also improve deposit quality, but must be enhanced by a carrier (detergent)  as
 the brightener is not water soluble. However, the carrier foams when agitated and
 with temperature increases. In the atmospheric evaporator, agitation of the zinc
 plating solution can send foam up the exhaust stack and onto the roof  Chanqinq
 from one vendor's carrier to a potentially less-foaming one can take months of
 production, as the  carrier is only removed by dragout of the zinc plating solution.

      Temperature increases  also affect brightener consumption and platinq qualitv
 Brighteners degrade in the plating solution is above 90°F. At such temperatures
 plating in recessed areas of a part deteriorates.

 Plating Trial With Reverse Osmosis Membranes

      In conducting the RO membrane plating trials, Global Plating used the following
procedure.

1.     Zinc  rinse tanks were dumped and made up with new RO water.


                                     132

-------
2.     The atmospheric evaporator and the RO system were activated.

3.     Samples of the RO reject, RO elutant, #1 rinse tank, and #2 rinse tank were
      analyzed for zinc, chloride, and conductivity. Zinc and chloride were analyzed
      by standard volumetric tests.  Conductivity was measured with a conductivity
      meter.

      RO membranes normally operate at about 80 percent efficiency  due to inherent
characteristic design and manufacturing parameters.  When the reject and elutant are
analyzed, the ratio between the two levels indicates efficiency separation.  The
efficiency expressed by this ratio also can indicate when the membranes are
becoming clogged and need to be cleaned. As the RO and UF membranes become
clogged, the conductivity ratio of concentrate to elutant and the conductivity ratio of
the RO concentrate to the concentration in the #1 rinse tank reduce.

COST OF DEMONSTRATION

      The total cost of the demonstration was $42,548.42. EPA provided $15,180
through the Pollution Prevention By and For Small Business Grant Program, and
Global Plating contributed the balance of $27,368.42.


RESULTS AND DISCUSSION

PERFORMANCE RESULTS

      The ratio of RO concentrate to RO elutant started at 17:1 and remained above
10:1 for over one month, as shown in Table 1.
TABLE 1.    RO Conductivity Ratios
Date
9-29-92
9-30-92
10-1-92
10-7-92
10-9-92
10-23-92
10-26-92
11-10-92
11-12-92
RO Concentrate
Conductivity
(ppm)
1,500
1,700
2,000
1,400
1,400
2,800
1,900
5,100
3,400
RQ Elutant
Conductivity
(ppm)
20
100
150
100
100
200
100
500
500
Ratio
Concentrate: Efut
ant
75:1
17:1
13.3:1
14:1
14:1
14:1
19:1
10.2:1
6.8:1
                                     133

-------
11-13-92
11-17-92
11-30-92
12-14-92
12-21-92
12-23-92
1-5-93
1-8-93
1-13-93
1-14-93
'' ===== !.
3,800
5,000
2,500
1,500
1,500
1,100
1,400
1,300
1,500
1,300
1
500
600
200
300
300
400
400
•— — ^ 	 _
300
400
500
1
7.6:1
8.3:1
13.5:1
5:1
5:1
2.7:1
3.5:1
	 	 	 	 	
4.3:1
3.7:1
2.6:1
'
 were
cleanfng.
membranes were becomin9
                                                                         and
 hona,1ch-      «e a?mosPhueric evaporator could not function efficiently
 because the rainy California weather resulted in 100 percent humidity with 35° to SOT

                                       the 6V« 
-------
CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      RO and UF membranes are effective in capturing zinc metals for recycling.

      Plating baths that operate at room temperatures use brighteners that are
destroyed at the high temperatures required for conventional evaporators.  A vacuum
still, operating at 90 to 100°F, would be effective for the evaporation of zinc chloride
baths.
      Zinc and nickel platers may find the process useful.  Nickel platers would  not
require the evaporation step, as nickel plating tanks, operating at 140° to 155°F,  lose a
great deal of water each day through  evaporation.  Other types of plating solutions
could be recovered using RO membranes providing they met membrane compatibility
requirements.

Barriers

      The atmospheric evaporator did not operate due to the high humidity during
the fall and winter months of 1992-1993.  The evaporator functions well in the summer
months (or in a dry climate). Because of this failure, Global Plating could not return
any concentrated zinc rinse solution to the zinc plating tanks, nor could they
demonsitrate any cost savings.

      The brighteners foamed  in the evaporator when the temperature approached
100°F. Research must  be conducted  to select a zinc brightener that can be used at
this higher temperature.

      As a potential solution to the evaporator malfunction, a RO and UF membrane
system could be placed in the zinc plating solution for water removal.  The
membranes would have to be tested to ensure they would function with the highly
concentrated chemicals in the plating  bath.
                                      135

-------
             IN-GROUND PLASTIC CONTAINER PRODUCTION SYSTEM
                TO REDUCE NITRATE AND PHOSPHATE POLLUTION

                                       by

                             Carl E. Whitcomb, Ph.D.
                                  Lacebark, Inc.
                               Stillwater, OK  74076
 ABSTRACT
       Lacebark, Inc. investigated the use of a new in-ground, plastic container system
 that reduced nitrate and phosphate pollution from above-ground container nurseries
 Such nurseries may use in excess of 2,000 pounds of nitrogen per acre per year-
 much of the fertilizer applied to containers may be lost through leaching and spillage
 The technique proposed by  Lacebark suggested that fertilizer rate could be reduced'
 by 50 percent or more (compared to conventional above-ground containers) when
 plants are grown in plastic containers that are submerged in the earth  In addition
 irrigation water demand could be reduced
 INTRODUCTION

 PROJECT DESCRIPTION

 Outline of Process/Product
       The in-ground plastic container production system is a new concept that may
 reduce nitrate and phosphate pollution from container nurseries. Container nurseries
 may use over 2,000 pounds of nitrogen per acre per year compared to 100 to 200
 pounds of nitrogen for conventional agriculture. Further, because conventional
 containers are above ground, they may tip or may be blown over, resulting in spillage
 of the soil and fertilizer and possibly damaging the plant. The temperature in an
 above-ground container may reach 100°F on a sunny day, causing slow-release
 fertilizers to release more quickly, and damaging roots such that their capacity to
 absorb nutrients is greatly  diminished.  Once the fertilizer exits the above-ground
 container, it is no longer accessible to the plant and becomes a pollution problem in
 soil and water run-off.                                               K

      With the in-ground plastic container system, the containers cannot blow over
 as the  containers are submerged in the ground up to the rim. Temperatures in the'
 container are typical of soil temperatures for the geographic area (73° to 78°F for this
 study). The cooler soil temperatures in the in-ground container improve the root's
 nutrient absorption efficiency compared to above-ground containers and with the
 unique container design, any nutrients that are leached from the container are still
 accessible by the secondary nutrient  recovery root system.

 Unique Product Features/Advantages

      The purpose of the project was to assess the practicality of growing plants in
specially-designed containers that are submerged in the soil.  This technique mav
reduce nitrate and phosphate pollution from containers, especially in sandy soils
                                     136

-------
where capture and recirculation of run-off water is not feasible as a pollution
prevention technique. Root zone temperatures in the in-ground container were at least
20°F cooler than in above-ground containers, thus improving root functions and
avoiding root death due to the high heat.

      The key element of the study was in-ground container that has over 1,200
openings, each 3/32-inches in diameter. Roots can  "escape" through these holes to
serve as a secondary nutrient recovery system; because of the small size of the
openings and the strength of the plastic, the roots cannot expand.  When the plant is
harvested, the roots break at the point of constriction, which has little effect of the
health and subsequent growth of the plant.

Product Drawing

      Lacebark developed and arranged for the manufacture of the new plastic
container design used in this project. The container is 12 inches in diameter and 10
inches deep, holding approximately 5 gallons of soil.  The sides of the container are
slightly  tapered to permit easy removal from the injection molder. The container has 4
ledges  on the sides, and each ledge and the container bottom  contain over 1,200
"escape" openings, 3/32-inches in diameter. The container has no large drain holes
as in conventional above-ground containers, as the large holes allow roots to escape
and grow to considerable size, making harvesting difficult and severely shocking the
plant if  any  roots must be severed.  The in-ground container is shown in Figure 1.
The pegs at the upper right aid in removal of the pot from the mold and prevent roots
from circling within the container.

Figure  1.    The In-ground Container
                                       137

-------
  APPLICATION

  Process or Products Replaced
        The in-ground container can replace conventional above-ground containers that
  are subject to being blown or tipped over, resulting in fertilizer spillage and possible
  plant damage. In-ground containers promote lower root temperatures, decreasing the
  rate of fertilizer release and increasing the nutrient absorbing capacity of the root
  system.
        Conventional field-grown nursery stock may also be grown in the in-ground
 containers, reducing fertilizer usage on a per acre basis. For example, when trees  or
 shrubs are grown in field soil  (without containers), the plant roots extend freely in all
 directions  Consequently, fertilizers are typically spread over the entire growing area
 some of which does not contain plant roots.  By growing plants in the field in in-   '
 ground containers, the fertilizers can be concentrated in the area where the roots
 develop. In addition with the "escape" roots and the secondary nutrient recovery they
 provide, nutrient pollution will be minimized and crop response maximized
 Wastes Prevented
       Fertilizer spillage from blown or tipped over containers is eliminated  Some
 slow-release fertilizers rapidly increase their rate of nutrient release as temperatures
 increase, as often seen in typical above-ground containers. With the in-qround
 containers  both plant roots and fertilizers are much cooler. This increases the
 SEfSLf ^l9*?0* to l£sorb nulrients while reducing the rate of release of some
 fn ™ nnnH38.6 ^zers;  The secondary "escape" root system that develops outside the
 in-ground container also may aid in reducing nutrient pollution in the soil and
 grounawater.
       The higher cost of the in-ground container is offset by its reusability as a
 production tool.  A plant can be grown in the in-ground container  harvested  and
 removed from the container when marketable size is reached.  The in-qround
 container can be reinserted into the same cavity in the soil for reuse.  The container in
 which the plant would be marketed would be a very thin, low-cost  conventlon™Dtesti!2
 container or a paper mache container that is plantable by the end-customer  The
 development of a paper mache container to complement the in-ground container has
 distinct advantages by reducing the number of plastic pots destinecI for landfills

       The direct  savings in fertilizer costs will be modest.  By confining the fertilizer to
 a small area, there will be secondary  benefits of reduced weed	"'-
 adjacent areas.
PROCEDURE

DEMONSTRATION
rhor.  I Lt  T    P Sltes ?*?* selected in central Florida for the demonstration
Cherry Lake Tree Farm in Lake County, west of Orlando, is on the deep roll nd  sand
mils generally referred to as the central Florida Ridge. The soils on this si e are'ver^
low ,n organic matter averaging 1.1 percent, and have a consistency of ban bearing.


                                      138

-------
Water and nutrient retention is quite low, and the first water table is at a depth of
approximately 30 feet, as estimated by the nursery manager.

      Loyd and Ring Tree Farm and Holly Lane Tree Farm are in eastern Hillsboro
County, approximately 20 miles east of Tampa, Florida.  Both sites are flat with a
shallow water table at a depth of approximately four feet.  The soils on both sites are
sandy, but the particle size of the sand is much finer in texture, with 3 percent organic
matter.  Water and nutrient holding capacity of these soils is substantially higher than
at Cherry Lake.

      Soil samples were taken from all sites before the assessment began.  The
analyses of these samples serve as a bench mark of soil nutrient conditions on each
site and are noted in Tables 4, 5, and 6.

      A tractor-mounted auger created the cavities in the soil into which  the
containers were inserted. Approximately 1.5 inches  of the container remained above
ground level. The container was filled with the existing sand field soil.  Young trees
were planted in the in-ground containers as well as directly in the ground (without
containers) in a random pattern and watered. Subsequent waterings were by drip-
irrigation. The plants were fertilized at a low, medium, or high rate according to
fertilizer label directions.  At Cherry Lake and Loyd and Ring Tree Farms, Osmocote
16-6-10 plus  minor nutrients with an 8 to 9 month release rate was applied at a rate of
100, 200, or 300 grams per plant.  At Holly Lane Tree Farm, Osmocote 14-14-14 with
a 3 to 4 month release rate was applied at a rate of 50, 100, and 150 grams per plant.

      Test plants at  Cherry Lake Tree Farm were sweet  gum, camphor tree, and
Muskogee crapemyrtle. At Holly Lane Tree Farm, live oak, American holly, and loblolly
pine were used as test plants. At Loyd and Ring Tree Farm, sycamore, citrus, and
red maple were used as  test species. All test plants were replicated four times with
the three levels of fertilizer and with and without the  presence of the in-ground plastic
container. 216 trees were used in  the study.

      Soil cavities for the in-ground containers were dug using a 12-inch tractor-
powered auger. A leg at the center of the auger functioned as a depth gauge.

       In Figure 2, the in-ground container is placed in the augered hole  prior to  filling
with soil.  The 3/32-inch holes may be seen, which allow roots to "escape" and
function as secondary  nutrient recovery systems outside the container. The four
groves  in the top of the pot hold the drip irrigation line in place to ensure accurate
watering.
                                       139

-------
 Figure 2.     The In-ground Container in the Soil
m««*hln 59iire*u3|*!u ln-9round container is removed from the soil after about 6
months.  Note that the roots have grown through the small holes in the container.

F:igure 3.    In-ground Container After Removal from the Soil
                                     140

-------
EVALUATION PARAMETERS

      The plants grew approximately four months at Holly Lane Tree Farm and for
three months at the other two site prior to growth measurements and further testing
for nitrates, phosphorus, and potassium. On June 13, 1993, height growth of all tree
species at Cherry Lake and Loyd and Ring Tree Farms was determined. Height
measurements were taken for the three test species at Holly Lane Tree Farm on June
17,  1993.

      After measuring height growth of the trees, the plants in the in-ground
containers were, lifted from the earth and a soil sample was taken from directly
beneath the bottom of the container.  For trees grown in the soil without the in-ground
container, a hole was  dug beside the tree to a depth equal to or greater than the
depth of the bottom of an in-ground container.  A large metal spoon was bent to
serve both as a depth gauge and to extract soil samples from beneath the plant at a
location similar to that described for the in-ground container.  54 soil samples were
taken and analyzed for nitrates,  phosphates, and potassium using conventional   .
agricultural soil test procedures.

      After taking the soil samples, the containers were returned to the cavities and
holes beside the non-container plants were refilled  so as not to overly disturb the
plants.  The plants will be allowed to grow for another year to obtain long-term
information.


ASSESSMENT SUMMARY

      The roots of  all species had grown sufficiently to extend through the 3/32-inch
"escape" openings in the in-ground containers and into the soil below, and to a lesser
degree, to the sides.  The roots outside the containers absorb water and nutrients and
send these to the leaves via the xylem, or central vascular strands, of the roots that
remain intact through  the small openings. The phloem, or outer root tissues, are
squeezed out by the expanding xylem, thus the downward flow of sugars from the
leaves in the phloem stops abruptly at the inner surface of the container.  As a result,
the roots outside the container are essentially nurse roots that capture water and
nutrients that otherwise would have been lost from a conventional above-ground
container, yet receive  very few sugars from the leaves in return. Because of root
constriction, the loss of these roots at a time of harvest has very little detrimental  effect
on the plant. Growth  of the trees (height in inches) generally was best with the lowest
fertilizer level or was approximately the same, regardless of fertilizer levels at all three
test sites, as shown in Tables 1, 2, and 3.  No significant differences in growth were
noted in any tree species between fertilizer levels or presence or absence of the
container.
                                      141

-------
 TABLE 1.   Plant Growth (Height In Inches) at Holly Lane Tree Farm, Plant City, FL
                                       FERTILIZER LEVEL
      SPECIES
LOW
MEDIUM
                                                                HIGH
  Live Oak
        Container
 9.6
  10.3
                                                                 9.1*
        No
        container
10.3
  11.6
12.1
  American Holly
        Container
 7.0
  6.3
                                                                 5.3
        No
        container
 5.7
  5.3
                                      4.7
  Loblolly Pine
TABLE 2.    Plant Growth (Height in Inches) at Cherry Lake Tree Farm, Groveland, FL
' ===========
SPECIES
Sweet Gum
Container
No
container
Camphor Tree
Container
No
container
=================«'-••— -—-—^—^-^^— —
FERTILIZER LEVEL I
LOW
MEDIUM
HIGH

7.7
6.3
6.1
4.5
4.5*
4.2

4.6
4.8
3.5
4.3
3.0
2.2
Muskogee Crapemyrtle
Container
No
	 container
* mean of four replica
8.0
7.7
:ions ~~
7.5
4.5
==================
6.2
3.5
==================
                                    142

-------
TABLE 3     Plant Growth (Height in Inches) at Loyd and Ring Tree Farm, Plant City,
            FL

SPECIES
FERTILIZER LEVEL
LOW
MEDIUM
HIGH
Sycamore
Container
No
container
11.5
11.5
9.2
11.6
9.6*
12.6
Citrus
Container
No
container
7.6
6.1
4.9
6.1
4.3
4.2
Red Maple, second flush of growth
Container
No
container
8.25
7.2
6.5
4.9
4.7
4.0
Red Maple, total growth (first and second flushes)
Container
No
container
11.6
11.0
15.0
14.0
10.0
11.5
  mean of four replications
 COST OF DEMONSTRATION

       The total cost of the demonstration was $33,072.85, with EPA providing
 $16002.17 through the Pollution Prevention By and For Small Business Grant
 Program,  and the balance of $20,852.85 contributed by Lacebark.


 RESULTS AND DISCUSSION

 PERFORMANCE RESULTS

       The in-ground container worked well at Loyd and Ring Tree Farm where the
 soils were finer and had more organic matter, and results were similar to a preliminary
 study  conducted in Oklahoma in 1991. At the Cherry Lake Tree Farm site with "ball
 bearinq" soil, some benefits  have resulted from reduced water movement through the
 soil during heavy rains.  Benefits from Cherry Lake were less clear.
                                     143

-------
 PRODUCT QUALITY VARIANCE

       The quality and consistency of the in-ground containers were excellent once the
 initial plastics fabrication problems were solved.  The variation in performance appears
 to be related to the soils and drainage conditions on the site rather than the product.


 CONDITIONS THAT IMPACT PERFORMANCE

       Clearly the Cherry Lake site was an extreme in terms of low soil organic  matter
 and extremely coarse, deep sand, yet there are hundreds  of thousands of acres of
 similar soil in central Florida.  The results suggest that this technique performs better
 on  sandy soils that  have greater organic matter and water-holding capacities  Much
 of the sand soils of New Jersey, South Carolina, North  Florida, and eastern Texas are
 similar to the Loyd and Ring Tree Farm site where the container performed as
 anticipated.

 TABULATION  OF DATA

      Soil test results were compared for sites with and without the in-ground
 container for each of the three nutrient elements: nitrates, phosphorus, and potassium
 The pre-test nutrient levels in the soil, both range and average, are noted to the right
 of the following tables.

      Soil test results from Holly Lane Tree Farm (Table 4) show no significant
 differences among nitrates, phosphorus, or potassium with or without the in-ground
 container at any fertilizer rate. The levels of nutrients in all  cases were similar to or
 lower than the soil test levels from before the treatments were applied  Soil nutrient
 levels were lower three months after the fertilizer was applied, suggesting that the
 slow-release fertilizer failed to release any appreciable nutrients. This  problem is
 known to occur from time to time, as consistency in the time-release coating thickness
 of the fertilizer can be  difficult to maintain.
TABLE 4.    Holly Lane Tree Farm - Soil Nutrient Levels below the Pot Depth after
            approximately Four Months
Nutrient
Fertilizer
Level
Container
Mean
Nitrate (ppm)
Low
3
5
4
4.0*
Mediu
m
3
1
4
3.0*
Hig
h
5
4
3
4.0
Phosphorus (
Low
341
326
341
336*
Mediu
m
329
372
338
346*
ppm)
High
326
326
341
331*
Potassium (p
Low
16
15
19
16.7*
Mediu
m
16
30
13
19.7*
pm)
High
14
15
21
16.7
*
                                     144

-------
No
container
Mean
Baseline
readings
(pre-
study)
1
1
3
1.7*
3
5
4
4.0*
2
4
2
2.7*
Range 8-13
Mean 8.8 ppm
332
335
355
340*
341
390
372
368*
343
381
346
353*
Range 350-381
Mean 364 ppm
15
21
15
17.0*
13
28
33
247*
14
79
18
37.0
*
Range 29-75
Mean 49.8 ppm
    t significant
      At Cherry Lake Tree Farm, nitrate levels in the soil below the bottom of the
container were not significantly different for the container versus no container at the
low and medium fertilizer levels.  However, at high fertilizer application levels, high
levels of nitrates were found below the in-ground container (24.0 ppm) versus a
comparable depth below the trees with no container (8.3 ppm).  Phosphorus levels in
the soil below the bottom of the containers also were not significantly  different for the
container versus no container  at all three fertilizer levels.  Potassium levels in the soil
below the bottom of the containers were  not significantly different for the container
versus no container at the low and medium fertilizer levels.  A significant difference
was seen in the high-rate application of potassium beneath the container (25.7 ppm)
versus no container (7.7 ppm). Data for  Cherry Lake Tree Farm may  be found in
Table 5.

TABLE 5.    Cherry Lake Tree Farm - Soil Nutrient Levels below the Pot Depth after
             approximately Three Months
Nutrient
Fertilizer
Level
Container
Mean
No
container
Mean
Nitrate (ppm)
Low
18
19
10
15.7*
5
18
8
10.3*
Mediu
m
12
23
7
14.0*
35
9
7
17.7*
High
26
16
30
24.0
t
9
4
12
8.3f
Phos
Low
54
54
31
49.7*
36
51
47
44.7*
phorus i
Medi
um
85
54
56
65.0*
62
49
51
54.0*
ppm)
High
56
62
51
56.3*
44
46
48
46.0*
Potassium (ppm)
Low
20
21
13
18.0*
5
22
10
12,3*
Mediu
m
19
48
21
29.3*
45
18
13
24.0*
High
38
23
16
25.7
t
8
8
9
7.7t
                                       145

-------
Baseline
readings
(pre-
study)

Range 5-13
Mean 9.3 ppm
Range 36-53
Mean 45.2 ppm
Range 8-21
Mean 1 1 .2 ppm
* Not significant
t Significant at 5% level
       At Loyd and Ring Tree Farm, nitrate levels in the soil below the bottom of the
 container were significantly lower beneath the in-ground container versus no container
 for the low and medium fertilizer level. No significant difference was seen at the high
 fertilizer level.  Phosphorus levels and potassium levels  (at all three fertilizer levels)
 were also not significantly  different in the soils beneath the container versus no
 container. Data from Loyd and Ring Tree Farm are shown in Table 6.
 TABLE 6.    Loyd and Ring Tree Farm - Soil Nutrient Levels below the Pot Depth after
             approximately Three Months
Nutrient
Fertilizer
Level
Container
Mean
No
container
Mean
Baseline
readings
(pre-
study)
Nitrate (ppm)
Low
8
6
9
7.7t
33
12
11
18.6f
Medi
urn
8
6
12
8.7f
21
9
17
15.7
t
High
10
7
17
11.3*
11
5
21
12.3*
Range 11-24
Mean 17.3 ppm
Phosphorus (ppm)
Low
40
33
56
43.0*
60
24
48
44.0*
Medi
um
91
26
68
61.7
*
41
25
38
34.7
*
High
47
39
83
56.3*
22
13
69
34.7*
Range 38-51
Mean 38.6 ppm
Potassium (ppm)
Low
6
11
7
8.0*
47
9
11
22.3*
r, ,.
High
12
6
9
9.0*
17
6
16
9.7*
9
12
21
14.0
*
11
5
33
16.3
*
Range 59-79
Mean 65.3 ppm
t Significant at 5% level
Conclusions

      A comparison of soil nitrate levels at a depth just below the bottom of the
container for Cherry Lake Tree Farm versus Loyd and Ring Tree Farm probably
                                       146

-------
reflects the differences in soils between the two sites.  The soil at Cherry Lake Tree
Farm was deep, "ball bearing-like" sand with an estimated depth of 30 feet to a water
table. The organic matter content averaged 1.1 percent.  By contrast, Loyd and  Ring
Tree Farm had a finer sand soil with much better water-holding characteristics and an
estimated depth to a water table of four feet.  The organic matter content averaged
3.0 percent.

      With the extremely well-drained sand at Cherry Lake Tree Farm and the heavy
rains that occurred during the period of the study, the container may have retarded
downward water movement and leaching compared to where no container was
present.  By contrast, at Loyd and Ring Tree  Farm that had a higher water table, and
where soils were also sandy, but of much finer texture and higher organic matter, the
benefits of the container fit the rough data of  the preliminary study.

      Phosphorus moves very little in soils, even in extremely coarse sand soils such
as at the Cherry Lake site.  The lack of difference with this nutrient between the
container and no container trials was not unexpected.

      Potassium moves at a modest rate under conditions of sand soils, low organic
matter, arid heavy rainfall.  A restriction was noted in potassium movement at the
Cherry Lake site similar to that observed for nitrates.

      The temperature of the soil at the time  of the second sampling two inches
below the surface averaged 78°, 73°, and 73°F for the Cherry Lake, Loyd and Ring,
and Holly Lane sites, respectively. On the same days while taking soil samples, the
soil temperatures in above-ground containers averaged 112°, 109°, and  104°F at the
same three sites, respectively. The temperature readings are the means of six
measurements.

COST/BENEFIT ANALYSIS

      Lacebark feels that because the in-ground container is so unique as compared
to conventional procedures, and since a full-scale production trial has not been
conducted, further cost/benefit estimates  would be mostly fiction.

      The unique in-ground containers performed well in terms of (1) reduced root
zone temperatures and corresponding reduced fertilizer release rates; (2) reduced
evaporation and water use; (3) avoided fertilizer loss, as trees could not be blown or
tip over, even though some were more than six feet tall; and (4)  reduced nitrate
movement. Benefits from this unique in-ground container will likely increase with
additional tree growth over time.

      The cost  of augering the holes to install the in-ground containers  is one-time for
a number of years of production. Whenever a tree is harvested, another container of
the same design and size would be inserted,  keeping the cavity  open.  The drip
irrigation system is also a one-time cost that will function for many years. The two
above items plus the cost of the in-ground containers are the primary up-front
expenses.,

      The in-ground container, as currently manufactured with a single cavity mold,
has a retail cost of $1.60, as compared to a volume-manufactured, blow-molded
                                      147

-------
 container that sells for $0.51.  Mass producing the container would reduce the
 production price drastically, lowering its selling price and improving its potential for
 acceptance in the marketplace.  The container cost can be spread over an anticipated
 life of four to six uses or more.


 CONCLUSIONS

 POLLUTION PREVENTION ASSESSMENT

 Incentives

       The design of the container and the root "escape" system was satisfactory
 Only minor modifications of the mold are required, specifically, a larger rib and more
 frequency around the inner circumference of the container to further reduce root
 circling.

       Placing the in-ground containers in the soil solves three major problems
 commonly noted with above-ground containers for nurseries and tree farms- heat
 injury to  roots in summer; cold injury to roots in winter; and blow over  The
 convenience of drip irrigation and resulting water savings are improvements over
 overhead sprinkler irrigation.

      The target industry consists of the many nurseries on sandy soils  currently
 producing plants in conventional above-ground containers.  Good soil drainage may
 be a requirement for this technique. This system may not work on sites  where water
 movement through heavier (such as clay) soils  is slow.  In-ground containers may  also
 have appeal to non-container field nurseries with sandy soils, which do not have
 adhesion qualities for good digging with mechanical tree spades.

      Additional assessments are planned for geographic areas where soils are
 appropriate for the use of in-ground containers.  As with any new concept successful
 demonstrations in similar growing conditions are usually necessary to promote
 technology adoption by nurseries and tree farm operators.

 Barriers

      Two major barriers are: (1) the initial cost of the container; and (2) in-ground
 containers are a  new technology.  Demonstrations  in various geographic areas can be
very effective in gaming familiarity with new technology. Constructing a mold such  that
four or more containers could be fabricated  at one time versus the  current sinqle
cavity mold would substantially reduce the unit cost to the end  user
                                     148

-------
                REUSE OF METAL FABRICATION WASTEWATERS
              VIA A NOVEL ULTRASONIC COALESCENCE PROCESS

                                      by

                               Dr. Scott R. Taylor
                           S.R. Taylor and Associates
                             Bartlesville, OK 74003
ABSTRACT

      Mertal working and finishing operations generate wastewater containing cutting
and cooling fluids along with metal particles and oils. Currently, this entire volume of
wastewater must be disposed of as a hazardous waste.  S.R. Taylor and Associates'
(SRTA) project studied the effectiveness of a novel ultrasonic technique that separates
the hazardous contaminants from the fluids so that the bulk microemulsion  phase can
be reused.  Up to 90 percent volume reduction of wastewater from small metal
fabricating shops was projected. SRTA built a multistage, multi-frequency operation to
assess trie commercial potential of the recovered oil phase and the solids sludge, and
to evaluate the long-term stability and usefulness of recycled fluids.


INTRODUCTION

PROJECT DESCRIPTION

      Typical metal working and finishing operations generate stable suspensions
containing hazardous solid metal particles as well as oils in their cutting and cooling
fluids.  Currently, the entire volume of this wastewater must be handled and disposed
of as hazardous waste.  The only acceptable methods to separate these  hazardous
materials are expensive and inefficient, especially for small operations.  The purpose of
this project was to assess the effectiveness of a novel ultrasonic technique that
achieves pollution prevention by separating the hazardous contaminants  from the fluid
so that the bulk microemulsion phase can be reused.  Successful implementation of
this process could feasibly lead to reuse of up to 90 percent of the volume  of
wastewater from small metal fabricating shops.  This project included the construction
of a full-scale unit for long-term quantitative and qualitative on-site evaluations of the
equipment, process, and reused fluid.

      Current metal cutting and finishing operations use skimming and filtration to
separate the wastes from their water-based cutting fluids.  The residual liquid from the
filtration consists of ultrafine metal  particles containing hazardous heavy metals such
as lead, zinc, and cadmium, and a highly stable emulsified oil-in-water phase.
Currently, the only accepted method of cleaning this waste mixture for  disposal  is
distillation, leaving a hazardous metal sludge, but allowing for the water to be safely
disposed,. Distillation is expensive, both for initial capital investment and for operation.
This expense deters many  metal finishers who, instead, simply filter the fluids and
attempt to recycle them. When  recycling is no longer possible, the entire volume  must
be disposed of as hazardous waste.  Other methods for separation of  oil from
                                      149

-------
 wastewater, such as air flotation, also have limited effectiveness.  They are not suitable
 for separation of multiphase Systems involving stable emulsions with suspended solids
 nor are they cost effective for small treatment volumes.  In all of these cases, the
 separation and removal of the small oil droplets and metal particles is so difficult
 because the particle size  is so small and becomes increasingly difficult as the particle
 size is reduced.  For example, the oil/water emulsion leaving an American Petroleum
 Institute separator normally has  oil droplets less than 30 microns in diameter and
 concentrations less than 200 milligrams per liter (Sylvester, N.D.and J.J.  Byeseda
 "Oil/Water Separation by Induced-Air Flotation," Soc. Pet. Eng. J., 20, p.  579, 1980).  A
 separation process is needed that will be effective on small particle sizes in small
 concentrations.

       The overall goal of this project was to further develop and optimize
 ultrasonically enhanced coalescence and separation of metal working fluid waste to
 allow reuse/recycling of the separated phases. The approach was to build a
 multistage, multi-frequency operation, to evaluate the commercial potential of the
 recovered oil phase and the solids sludge, and to conduct actual site tests to
 ascertain  long-term stability and usefulness of the recycled fluids.

 Theory

       The theoretical basis for separating a suspension of fine particles  involves the
 use of ultrasonic  waves transmitted into the bulk fluid in a standing wave pattern. This
 causes small particles to collide  with the large particles and thus coalesce to a size
 large enough to enable separation from the bulk fluid.

       Two theories have been developed that treat  the covibration of suspended
 particles in a fluid medium. The relative motion, R, between the particle and the
 medium, due to ultrasonic vibration, is a convenient  means of expressing the
 phenomenon. The  Brandt, Freund, and Hiedemann derivation (Brandt, O., et al   "Zur
 Theorie der Akustischen Koagulation," Kolloid-Zeitschrift, 77, 103, 1936) leads to nearly
 the same  mathematical expression given in later work by St.  Clair (St Clair  H W
 Industrial Engineering Chemistry, 41, 2434, 1949).

      The Hiedemann form is:
                             Xp =
                             Xm
where
Xp, Xm      = amplitudes of motion of particles and medium, respectively (cm)
d            = particle diameter (cm)
f            = frequency (Hz)
Pp.  Pm       = densities of particle and medium, respectively (g/cc)
r)            = viscosity of the medium (poise: g/cm/sec)
                                      150

-------
      Calculations based on the Hiedemann equation (taking into account the density
and viscosity of water to show the variance in R as a function and particle size)
indicate that within the frequency range of 50 to 100 kilohertz, it should be possible to
obtain a difference in the relative motion of solid particles and emulsified liquid
droplets (5 to 50 micron size) as long as the density difference between the  emulsified
phase and the continuous phase is 0.1  grams per cubic centimeter or greater.
Therefore, theoretically it is possible to coalesce oil and metal particles in a water-
based medium.

Unique  Process Features/Advantages

      The unique feature of this process is its ability to allow the  recycling of much of
the fluid used in metalworking  processes since it contains up to 90  percent water.
The process  can be operated as a simple, inexpensive service provided by outside
vendors already working with disposal o'f these fluids.  This reduces the cost to the
consumer while maintaining close control on hazardous wastes in the manufacturing
plant.

      Previous research under a Small Business Innovative Research Grant verified
the relationship of particle size density and frequency of operation.  The results of
these studies proved the "soluble oil"/water microemulsion is the bulk phase and the
coalescence  method effectively removes the excess oil and larger solids from this
phase.  This leaves a  cleaned bulk  phase that could be 100 percent reused.

      Particles below 15 microns were not expected to be greatly influenced by the
60 kilohertz vibrations used in these earlier studies. Also, with oil present in  the
system, the viscosity increases slightly,  and the oil may coat the suspended  solids. At
a higher frequency (100 kilohertz), solids separation should improve.  However, much
lower flow rates will be required to account for the higher attenuation at the higher
frequency. Better solids rejection should be achieved by using a staged coalescence
with two chambers, one at 60 kilohertz, and a second at 100 kilohertz.  Manufacturing
and testing of this multi-stage array were the main purposes of the SBIR project.  In
addition, a portion  of the reconditioned fluids was returned to the operators to learn
whether the cleaned fluid was suitable for reuse.  Each sample from the earlier studies
proved  satisfactory for reuse and did not seem to go rancid as quickly.  Verifying
operator reuse was also part of this project.

Process Schematic

      A schematic drawing of the process is shown in Figure 1.  As can be  seen, it is
a simple, continuous flow system that allows any volume of fluid to be treated.
                                      151

-------
Figure 1.     Schematic of Continuous Flow-through Array
                           0)
                           .a
                                                       O>

                                                       33
                                                       o
                                                                        .
                                                                       E P
                                                                       to Q-
                                                                       CO
                                                             Q)
                                                             .a

                                                             E
                                                             (0

                                                           £6

                                                           -* 
-------
       The coalescence chamber contains a flexural plate developed by SRTA.  When
 power is applied to the system, the plate radiates a sound beam in a circular area.
 The axisymmetric quality of the sound beam is very effective in causing the smaller
 partjcles in the liquid to collect together as larger particles, or coalesce. The larger
 particles then separate into different phases by gravity and differing densities.

       When a stable suspension of metal cutting fluid waste is pumped through this
 coalescence chamber and the phases are allowed to separate, a water/"soluble oil"
 microemulsion remains between the oil and metal sludge phases.  This microemulsion
 may be removed and reused as a  metal cutting fluid.  The top oil layer may also be
 removed as a machine-ways oil or recycled. The bottom metal sludge layer may be
 sold for metal values, or disposed  of properly.

 Implementation in Industry

       Based on results from the process testing, specifications can be developed for
 an actual system.  In the tested scenario, the system can be envisaged as a service
 operation much like the current parts cleaner service systems.  A cooling fluid holding
 tank (50 to 100 gallons capacity) could be installed at the customer's facility.  The
 coalescence equipment would be mounted on a portable skid. The service operator
 would visit the customer's facility, treating the cooling fluids, and pumping the treated
 fluids into the holding tank where settling would occur.  As the customer needs
 cooling fluid, he removes it through taps on the side of the holding tank.  Periodically,
 the separated top oil phase  could be collected and either used internally at the facility
 or sold to an oil recycling firm.  The sludge, which settles to the bottom of the holding
 tank, would need to be removed and sold for recovery of metals; as a  final  option,  it
 must be disposed of as hazardous.

       The cost for the  proposed system is  minimal since the only  operating expense
 is for electricity.  Capital costs are relatively  low due to the simplicity of the system.   To
 date, S.R. Taylor has successfully treated liquids at flow rates up to 10  gallons per
 minute. Large volume treatment will be handled using a modular approach with
 several units operating  in parallel to provide the  desired treatment  rate. Also, large
 operators may prefer to have the equipment installed on site to allow more frequent
 operation than could be reasonably handled by a service company.

       From the customer's point of view, no water  is discharged, the valuable oil is
 recovered, and the hazardous sludge is handled by specialists.  By reusing the bulk
 microemulsion phase, the oil and metal values are essentially concentrated.  If this
 concentration is sufficient, these materials represent new resources rather than
 hazardous wastes. The overall goal is to reuse/recycle all materials contained in
 contaminated metal cutting fluids.

APPLICATION

 Process/Products Replaced

      Current practice  requires that metalworking facilities either dispose of the used
fluids as hazardous waste or treat the fluids on  site  to completely separate the oil,
solids, and water phases.  Even then, the oil and solids phases need to be  handled as
                                      153

-------
hazardous wastes.  Systems designed to provide such complete separation are
expensive and not easy to implement for small operators.

      Replacement of these expensive processes with a simple service that allows
most of the cooling microemulsion to be recycled will greatly ease the burden of
compliance on metalworking shops, particularly small operators.

Wastes Prevented

      Industries that discharge their wastewaters directly to streams, rivers, and lakes
have to comply with limits set by the U.S. Environmental Protection Agency and their
state on the contaminant content of their effluents. Typical  contaminants may include
oils and grease; oxidation and biodegradation products; and colloidal particles of
heavy metals.  Such effluents come from such sources as refineries; de-inking
operations; metal plating; iron and steel plants; chemical processing plants; barrel and
drum cleaning; washrack and equipment maintenance; mill  waste; and aluminum
forming.

      Of these industrial wastewaters, metal fabricating processes have some of the
highest oil contamination concentrations as  seen in Table 1. For this reason, these
fluids have been selected for initial testing although the proposed method should have
applicability to a large number of these wastewaters. Each year, approximately 260 to
800 million gallons of machining fluids are used in the U.S.  When used, these "soluble
oil" and water fluids tend to form stable suspensions with hazardous metal particles.
Removal of the excess oil and solids from the  suspension would  allow the remaining
water/"soluble oil" fluid to be reused.
TABLE 1.    Typical Oil and Grease Contents
Source
Range of Oil/Grease Concentrations
(percent)
Wastewater
Sewage
Food Processing
Textile
Petroleum Refining
0.001
0.01
0.001 -
0.01
-0.01
-0.1
0.005
-0.1
Primary Metals
Rinse Waters
Concentrate
Metal Fabrication
0.001
1 -
1 -
-0.1
5
15
Metal Cleaning
Rinse Waters
0.001
-0.1
bource:  Kulowiec, J.J., Pollution hngmeermg, 11,49, 19/9.
                                      154

-------
Cross Segment Uses

      Other industries potentially benefiting from this separation technology include
chemical processing for separation of liquid-liquid emulsions, the oil industry for
cleanup of process waters from enhanced oil recovery operations and for the
improvement of inertial separators, and food processing plants for wastewater
treatment.

      While the technology has application to a wide variety of separations, it is a
pretreatment for use prior to other separation techniques, such as gravity settling.
Hence, the technology can be usefully applied to any system that has two insoluble
phases that  have a sufficient specific gravity difference to allow the ultrasonic standing
wave to set up relative motion differences between the particles.  This leads to an
increase  in particle size distribution that allows easier separation.


PROCEDURE

DEMONSTRATION PROCEDURE

      The following objectives were set as the next logical steps toward refining the
coalescence separation process. Successful completion of these objectives should
provide answers to the  questions listed below.

1.     Quantitatively evaluate effectiveness for providing a usable fluid over extended
      periods.

      •     How long  (i.e., how many cycles) can the emulsion be reused?  What
            percentage of the material is reusable following a cycle? Is this
            percentage a function of time/number of cycles? What other variables
            may affect this percentage?

2.     Qualitatively evaluate process effectiveness including determining seasonal
      (summer versus winter) effects  and obtaining operator feedback on reused
      fluid.

      •     Operators have complained that cooling fluid becomes rancid during
            summer months.  In previous testing, metal operators noted that the
            rancid  odor disappeared after separation.  Will this continue to be the
            case after  continued reuse? How does the reused fluid compare to new
            fluid?  Does anything need to be added to the emulsion before reuse?

3.     Determine economic advantages for a metal operator using this process.

      •      What are the commercial values of the oil and metal sludge layers?  Will
            the oil need to be recycled or can the separated oil be used as is?  Can
            the metal sludge be sold as a resource for metal values, or will it need to
            be handled as a hazardous  waste?

      The following work plan was  proposed to meet the above objectives.


                                      155

-------
 1.     Construct portable multistage, multi-frequency array.  The objective of this
       task was to construct a multistage, multi-frequency system to enhance the rate
       of separation by incorporating quiescent zones into the pump-around loop.
       Based on prior testing, improved solids coalescence can be achieved by
       operating at a higher frequency up to 100 kilohertz.  However, increasing
       attenuation  effects limit the flow rate, thereby limiting the treatment rate at these
       higher frequencies. The proposed solution was to conduct a simultaneous
       multi-frequency operation. This design allowed operation at rates up to 10
       gallons per  minute while treating a portion of the stream at 100 kilohertz at a
       flow rate of  2 to 3 gallons  per minute. After assembly, the system was tested
       with simulated fluids to verify proper operation.

       Since both streams are recombined in the holding tank, the entire volume
       eventually passes through both treatment chambers as long as the fluid is
       treated for a sufficient period to allow several passes through the slower
       chamber.

       Ceramic crystals from Piezo  Kinetics, Inc.  were used to build the 60 and 100
       kilohertz transducers for the  two treatment chambers in the multistage array.

       A magnetic-drive  chemical pump with a Ryton pump housing was purchased
       for the portable array from Little Giant Pump Company. The pump handles a
       maximum pressure of 12.7 pounds per square inch at a maximum rate of 20.0
       gallons per  minute. The pump uses a one-eighth horsepower engine, however
       actual energy consumption by the pump and the entire array was measured
       throughout the testing.

       The completed  multistage  array was mounted on a wheeled skid for ease in
       transportation.

2.     Install holding  tank on site. The on-site  service array consisted of a 50-gallon
       separation tank that was installed at Delta Manufacturing,  a stainless steel
       fabricator. The  tank was made from plexiglass to allow observation of fluid
       separation, and several taps  were located along the side of the tank to permit
       removal of "cleaned" fluid for reuse. The tank also included a tap on the
       bottom for periodic sludge removal. The tank was constructed with a
       removable top to  allow for adequate cleaning prior to removal from the test site.

       Delta chose  to use the tank for storage of fluids from a grinder and a saw. The
       resulting 20  gallons of dark grey fluid constituted the test sample.  This material
       normally would  be disposed  at this stage of contamination. Delta planned  to
       replace this  fluid with fresher solutions from a lathe; however, this did not occur
       in the time allotted for the testing, so the initial dirty sample was the only
      sample treated.

3.    Treat suspension at regular intervals through varying weather conditions.
      On site testing required periodic visits to treat the fluids to promote settling and
      separation of the individual phases. The schedule called for treatments every
      two weeks throughout the year.  The actual schedule is provided in a later
      section.
                                      156

-------
4.    Interview metal operators using recycled fluid.  The test site operators were
      interviewed to obtain an evaluation of the fluid after long periods of recycle (i.e.,
      how does the fluid respond after 2 recycles?, 5 recycles?).

5.    Perform composition tests of the clarified water layer. The goal of this task
      was to determine the level of oil, grease, and total suspended solids (TSS) in
      the water phase for recycle.  These analyses were done by an outside testing
      laboratory after all test samples had been taken. Oil and grease content was
      measured using EPA method 413.1 and TSS was measured using EPA method
      160.2.
EVALUATION PARAMETERS

      SRTA designed a storage tank, which was built by General Plastic Fabricators
of Tulsa, Oklahoma, for on-site installation.  The tank was fabricated from transparent
acrylic for maximum visibility of the fluids throughout the testing process. The tank
was 4 feet, 8 inches from the top to the base plate, 12 inches wide, and 18 inches
deep, for a volume of approximately 50 gallons. The top and base of the tank were
flanged  1 inch on all four sides and are connected with a series of bolts to allow
removal for easy cleaning. Two holes were drilled in the top plate for insertion of
thermometers to monitor temperature changes from treatment to treatment, and
before and after treatment.  The base plate was molded into a sloping funnel shape
with a central 2 inch plug for removal of metal sludge.

      The tank was filled with water and checked for leaks after 12 hours.  No
moisture was noted at any joints or around the base of the tank.  Due to the weight of
the water and the pressure exerted on the walls of the tank, supports were added on
all four sides of the tank, 10 inches and 24 inches from the base plate, for extra
strength.

      On the side of the tank, six, one-half-inch diameter spouts were placed 9 inches
apart, starting 2 inches from the base plate continuing to 6 inches from the top. A  3-
inch wide strip of acrylic was placed under the spouts for reinforcement.  These
spouts were for removal of selected layers of the separated fluids.

      Delta received the appropriate soluble oil for use with the grinder, and put a
sample of the wastewater from the grinder sump into the holding tank in late October
1992. This represented the worst fluids generated by Delta, as the material was ready
to be disposed of; no other method was used to try to recycle this fluid.

      Delta continuously added used fluids through the top plate.  Every two weeks,
the fluids were pumped from the tank through a hose connected 8 inches from the
base plate.  The fluids were pumped through the multistage, multi-frequency array into
the coalescence unit; they were returned to the tank through the top plate to settle
into layers.

      The water was very dirty with a dark grey color, and had apparently
decomposed significantly.  SRTA advised Delta that it was unlikely that the treatment
could clarify this wastewater. Delta was more concerned that the fluid could be
                                      157

-------
 cleaned sufficiently for reuse.  (As noted below, this became an issue of user opinion
 rather than analysis.)

 TESTING

      The sample liquid was prepared for testing as described above. Additional
 fluids were added to the sample midway through the program.  This waste sample
 was treated over three months as listed in Table 2.
TABLE 2.   Actual Schedule for Coalescence Treatment
Treatment #
1
2
3
4
5
6
7
Cumulative Time, weeks
0
2
3
5
8
11
12
Date
November 6, 1 992
November 16, 1992
November 25, 1 992
December 9, 1 992
December 29, 1 992
January 25, 1993
February 4, 1 993
      Testing consisted of attaching the treatment array to the holding tank, pumping
the fluids out of the tank, through the two treatment chambers, and back into the tank
in a continuous flow cycle. The ultrasonic power generator activated the plate(s)  The
initial treatment on November 6, 1992 was done at 60 kilohertz only; the 100 kilohertz
plate was not operational due to an electronic failure in the power generator.  All
subsequent tests were conducted with  both plates activated. The individual test dates
are described below.

      November 6, 1992: This initial test verified that the fluids could be pumped
      through the treatment array and  verified that the 60 kilohertz plate was  running.
      Samples were taken before and  immediately after treatment  (approximately 30
      minutes). The fluid has a fairly strong, rancid odor.

•     November 16, 1992: No fluids were added or removed from the tank.
      Treatment was the same as the first test.  Two samples were taken: one prior to
      treatment and one 30 minutes after treatment. The return water stream, which
      entered the holding tank vertically through the top plate, caused significant
      agitation of the remaining fluid in the holding tank. Since this could cause
      resuspension of already-settled solids, this was modified prior to the remaining
      tests.

      November 25, 1992: No fluids were added or removed from the tank
      Treatment was similar, but the return line to the holding tank was repositioned
                                     158

-------
      to allow the water to flow in horizontally before dropping into the remainder of
      the water in the tank. This greatly reduced the agitation of the water.  Two
      samples were taken: one prior to treatment and one 30 minutes after treatment.
      The oil and solids content had been reduced.  However, a rather large
      algal/bacterial growth had formed on the top oil layer.  These cultures exist in
      all shops and are a serious deterrent to fluid reuse.  The fouled fluid has a
      noxious odor, and was separated prior to the next test. The water itself does
      not have  a bad odor and, after treatment, has the smell of oil, not the  bacterial
      byproducts. The color of the fluid is still a dark grey. This color, along with the
      growth on the  oil layer, has inhibited the personnel from recycling the  water
      phase.

•     December 9, 1992: A skimmer was used to  remove the top oil layer that had
      significant algal and bacterial growth  and a foul odor.  No fluids had been
      added or removed.  The need to recycle some of the fluid was discussed with
      Delta personnel. Again, two samples were taken: one prior  to treatment and
      one 30 minutes after treatment.

•     December 29,  1992:  Two gallons of additional fluid from the grinder had been
      added to the tank, although none of the fluid had been recycled.  Again, two
      samples were taken: one prior to treatment and one 30 minutes after treatment.
      Visual observation of the two samples showed the separation easily.  The
      treated sample was much lighter in color and clear separation was seen.
      However, even with the enhanced separation, the fluid remained a dirty color,
      and this appeared to inhibit personnel from recycling the water phase even
      though it was obviously cleaner than the starting material.

•     January 25, 1993:  Delta planned to recycle some of the fluid before the next
      treatment. After removing several gallons of the water phase, the odor was
      determined to  be too strong to allow recycling.  SRTA believes that the liquid
      was allowed to be stored too long and should have  been recycled much
      sooner. This delay permitted bacteria and algal cultures to overwhelm the fluid.


      The fluid  was treated for the last time and sampled again. Although the water
      still smelled strongly at times, the operators did agree that it was no worse than
      the initial material. This means that the treatment was effective  in controlling the
      fluid composition even though this particular sample was too fouled at the start.

•     February 4, 1993: Delta indicated that they would replace the fluid with material
      from a lathe, but this had not yet been done.  (The lathe fluid never gets as
      dirty as the grinder fluid, so it has a higher potential  for recycling.)  One final
      sample was taken.

ASSESSMENT SUMMARY

      Delta Manufacturing did not recycle any  of the sample nor did they fulfill their
promise to provide a fresher sample.  This limits the usefulness of these tests.
However, future development can proceed quickly,  as SRTA has a complete, mobile
system that can be applied for further testing.
                                      159

-------
       Although the specific goals of this project were not fully accomplished due to
 problems with the availability of test samples, SRTA has achieved a significant step
 towards commercialization by providing the first long-term data and by providing a
 system for additional testing.

       As a result of having conducted this project, SRTA presented the results at
 several conferences including the EPA and DOE Waste Stream Minimization and
 Utilization Technology Fair,  April 22-23, 1993, Austin, Texas.


 RESULTS AND DISCUSSION

 PERFORMANCE RESULTS

       Technically, the system operated correctly and SRTA obtained  similar
 separation results to those observed in earlier studies.  Unfortunately, the actual
 sample used for these tests was an extremely contaminated sample and even
 operation at 100 kilohertz did not product sufficient visual separation to satisfy the
 user, although the analytical results were very good.  Table 3  shows a comparison of
 the analytical results from an earlier sample and the current sample.
TABLE 3.    Ultrasonic Coalescence Data for the Actual Cooling Fluid Samples

Oil & %
Grease, Decrease
mg/L
TSS, mg/L
% Decrease
Sample A - Prior Study
Initial mixed sample
Ultrasonic, middle layer
Sample B - Current Sam
Initial mixed sample
Ultrasonic, middle layer
26,456
17,656 33
757
551
pie
27,800
10,200 j 61
27

3,430
980
72
      The first difference in these two samples is that Sample B (in the current study)
had a much larger TSS than the earlier sample.  As a result, although both the oil and
grease and TSS levels were greatly reduced, the TSS still was larger than  in the earlier
sample.  However, the earlier sample was entirely suitable for recycling.  Although oil,
grease, and TSS content were reduced sufficiently to allow successful recycling, the '
dark color of the treated fluid (presumably resulting from the high initial solids content)
was an impediment to user comfort. Thus, none of the fluid in the current study was
actually recycled.

      Clearly, the process has some limitations on the quality of fluid that can be
accepted for treatment.  If the fluids are badly contaminated at the start  even good
analytical results will not convince the user to recycle the fluid.         '

                                      160

-------
PRODUCT QUALITY VARIANCE

      While separation was better than that achieved in earlier tests, resujts for TSS
were somewhat disappointing.  SRTA had anticipated greater solids rejection with the
addition of the 100 kilohertz chamber.  Two reasons are suspected for this product
quality variance.

      First, the starting fluid that was taken from a grinder sump tank was already at
the point at which the operator would normally discard  it.  The  fluid had already
reached a level of deterioration that would not normally be considered  "cleanable" as
seen by the very dark color and the high TSS of the suspension prior to treatment.
This material probably had a high concentration of extremely fine particles, and such
fine material may not have been easily coalesced even  at 100 kilohertz.

      Second, although SRTA tested the system at both 60 and 100 kilohertz before
using it to treat the fluids, a problem was discovered with the power generator after
completion of the first two treatments: the 100 kilohertz section may not have been
fully optimized. SRTA made modifications to the system that improved its
performance, but this may have affected  the initial results.

CONDITIONS THAT IMPACT PERFORMANCE

      Clearly, the higher the solids loading, the poorer the performance of the system.
These microemulsions can act much like thixotropic fluids to stabilize suspended solid
paniculate.  Further testing with the 100 kilohertz section alone should  be conducted
to determine maximum tolerable solids content and to further optimize  the system to
treat high solids content fluids.

TABULATION  OF DATA

Testing Observations

      The fluid was so dirty at the beginning of the test that it  represented a worst-
case test. Since the treatment was not intended to provide a complete separation, the
resulting fluid,  although clean enough to  recycle, was not suitable to the users'
subjective judgement. This is an important factor leading to commercialization.
Analytical characterization can verify separation and improve user confidence in
recycling. The first requirement for the user would be to start with reasonably fresh
fluids,  the process then would operate to maintain fluid cleanliness.

      The basic treatment procedure was  simple to conduct.  Hookup required simple
connection to  two hoses, and a single 110 volt outlet can provide power.  All
treatments lasted 30 minutes. Total time required to connect the system, treat, and
clean up was less than one hour.

Analytical Results

      Analytical results for oil and grease analysis and TSS are listed in Table 4 and
are shown graphically in Figure 2. The fluid composition was greatly affected  by the
treatment.  Reductions  in oil and grease  content reached 61 percent, while 72 percent
                                      161

-------
reduction in TSS was achieved. These results were better than those obtained in the
earlier studies on cleaner feed stocks.
TABLE 4.   Analytical Results for Treatment Samples
Treatment Number
1 (initial sample)
2
3
4
5
6
7 (2 gallons added)
8
9 (5 gallons added)
10
11
Oil & Grease, mq/L
27,800
15,600
12,400
12,000
11,400
10,800
14,700
14,300
21,900
12,900
14,500
TSS, ms/L
3,430
1,500
1,120
1,050
1,050
980
2,180
1,760
1,780
1,740
1,270
      Figure 2 also shows when additions were made to the total fluid, and it can be
seen that the treatment works well in providing a stable,  consistent composition in the
cooling fluid.  Even though the analytical results looked promising, the fluid was
deemed non-recyclable by the users because it "looked  bad." Obviously, some
method of either achieving a better appearance is required or the user needs to be
made aware of the analytical results immediately in order to increase the likelihood of
fluid reuse.
                                      162

-------
Figure 2.    Microemulsion Fluid Composition as a Function of Time and Treatment
            (period dates indicate sample numbers)
               0

-r
\



-*;
- -
\
V
\

^-
— •- Oil/Grease, mg/L
-<•- TSS, mg/L

- -




"^»

- *
- • -


-*.

A
- -


-»-

^
• - -


V

_^^
- - -




V
- -

/
-•

V

- - -
R
/

\


\
- - -

\
V

"i
. . _


^"

-•
                   1   2    3

                (11/6/92)
4   5    6   7   8   9   10   11

    Sample No.
                             (2/4/93)
                                     163

-------
 COST/BENEFIT ANALYSIS

      A representative from a local (Tulsa-based) waste disposal company was
 interviewed for information regarding metal fabrication shop wastewater disposal
 costs. To begin service with this company, a one-time start-up fee of $250 is charged
 for analysis of a waste sample to characterize the waste and to determine optimum
 disposal methods. The charge for disposal is $180 per 55-gallon drum.  A pick-up
 charge of $130 is incurred each time the disposal company removes the wastewaters
 regardless of the number of drums. This encourages the small business to store
 large quantities of wastewaters on-site to avoid high hauling charges.  The disposal
 company representative reported their average customer has wastewaters hauled 2 to
 5 times per year.

      The owner of a local, small (15 employees) metalworking shop  was also
 interviewed regarding wastewater disposal costs. His company disposes of
 approximately six 55-gallon drums annually, and he reports the disposal costs are
 "very high." He is pursuing the possibility of purchasing an on-site recycling system at
 a potential cost of $23,000. Clearly, a need exists for an inexpensive alternative for
 disposing of metalworking wastewaters by small businesses.

      Based  on the charges suggested above, a small operation disposing of three
 55-gallon drums (1 drum 3 times) could incur the following costs as shown in Table 5.
TABLE 5.
Cost Comparison of Disposal of Metalworking Wastewaters and
Coalescence Treatment for Recycling
Disposal
One-time analytical fee
Pickup charge (x3)
Disposal charge (x3)
Total
$250
$390
$540
$1,180
Anticipated costs for SRTA service
Operating costs
Amortized equipment cost (~ $40,000; 7 years
with 1 5 customers/month/year)
~$0.45/gallon/treatment
~$32/customer/treatment
For treatment of a 25 gallon sample (10 per year)
Operating cost (0.45 x 25 x 10)
Capital cost (32 x 10)
Profit @ 15% (0.15 x 432.50)
Total
i 	 =====^==========================5===B=5
$112.50
$320.00
$65.00
$497.50
======= —
                                    164

-------
      This represents a savings of almost 58 percent to the user.  The process could
provide a real economic benefit to small users if it enables them to avoid disposing of
their wastewaters.  Even if SRTA's operating costs double (possibly from higher
wages, longer treatment times, and the like), a significant savings over disposal could
still be realized.


CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      The user can pay a small monthly service charge to have  his wastewaters
treated, thus allowing him to recycle virtually all of the water.  The incentives are lower
cost, less potential for accidental discharge or mismanagement of hazardous wastes,
and less problems associated with meeting state and federal regulations.

Barriers

      The most immediate barrier is that although the fluid may  be analytically clean
enough to recycle, the operator, through lack of knowledge, may choose not to
recycle due to visual quality. This can be overcome by providing adequate analytical
results at the time of treatment to verify separation in order to build user confidence.
This process is less effective on heavily contaminated fluids, since it does not  provide
complete separation.

      A secondary barrier is a lack of experiential data to show to potential users.
Continued development and field testing, which is being conducted, will overcome this
problem.
                                      165

-------
          REDUCTION OR ELIMINATION OF COOLING TOWER CHEMICALS

                                       by

                         Larry Stenger and Thomas Dobbs
                          Water Equipment Technologies
                           West Palm Beach, FL  33411
 ABSTRACT
       Cooling towers used in air conditioning applications are a familiar site near
 many buildings. Since cooling towers are vast heat exchangers, the water in the
 tower becomes a perfect incubator to grow bacteria, algae, and fungi.  Cooling towers
 are also plagued by corrosion, scale, and sediment accumulation. Water Equipment
 Technologies (WET) addressed these problems through an innovative process utilizing
 a zinc and copper alloy that is placed in the cooling tower sump.  The reaction of the
 cooling tower water with the alloy raises the water's pH, and thus reduces corrosion
 prevents algae from forming, and destroys bacteria.  This concept reduces or
 eliminates the need for chemical algaecides, fungicides, corrosion inhibitors  and  scale
 inhibitors currently used to maintain cooling towers.
 INTRODUCTION

 PROJECT DESCRIPTION

 Background
       In 1987, Don Heskett, an inventor, was experimenting with the removal of
 chlorine from water and inadvertently used a piece of brass to stir chlorinated water in
 a glass beaker.  To his amazement, the chlorine disappeared. The brass consisted
 primarily of copper and zinc with smaller quantities of lead, iron, and cadmium  Since
 lead and cadmium are undesirable metals in water supplies, a special alloy of zinc and
 copper - named KDF -- was formulated to achieve the same results.

       Further tests  have shown that a mixture of highly pure, molten zinc and copper
 yields an alloy with millions of bimetallic couples.  The zinc acts as the anode and
 copper acts as the cathode. Exposing sufficient surface area of the alloy to water
 creates an electrical potential (Eh). The Eh for zinc is -0.76 milli-volts, and the Eh for
 copper is +0.34  milli-volts. Zinc becomes the electron donator and the resulting
 redox reaction -- corrosion -- causes electro-chemical reactions that provide a "natural"
 treatment process to control the formation of hardness scale, bio-film development
 chlorine removal, and ionic heavy metal reductions.

      KDF comes in several forms (e.g., granules, filament or wool, wire or brush and
 powder) with varying levels of pure copper and zinc,  depending on the application
The granules are used in backwashing-type filters much the same way sand is used to
filter water, except that KDF also removes chlorine, bacteria, and heavy metals  The
wool is used on recirculation loops, lending itself to applications such as cooling


                                     166

-------
towers.  The brush form is for small portable devices where the removal of chlorine
after disinfection is desired. The powder provides faster reaction times in treating
wastewater (and is currently under development).

The KDF Process

      The KDF process is as follows:

1     Chlorine is removed through its conversion to zinc chloride (Cu/Zn + CI2 =
      Zn++ + 2CI~).  Zinc loses 2 electrons through oxidation, and chlorine gains 2
      electrons through reduction.

2     Heavy metals are removed by plating-out the metals onto the cathode  (copper)
      sites (Cu/Zn  + Pb++(NO3)2. = Zn/Cu/Pb + Zn++((NO3)2.). Zinc loses 2  electrons
      through oxidation, lead gams 2  electrons through reduction and plates-out on
      the copper sites. Excess zinc ions have a tendency to redeposit back onto the
      media.

3     Calcium carbonate precipitation and scaling are controlled  by an undetermined
      mechanism that reduces the calcium carbonate's ability to  link together.  The
      electrochemical reactions interfere with the crystalline structure of limescale, and
      a powdery, rather than vitreous, scale formation is seen when the water dries in
      the splash areas of the cooling  tower.  No scale  is noted on the heat transfer
      surfaces, and no chemical formula has been developed to  explain this
      phenomenon.

4.    Bio-film accumulation and bacterial growth are controlled by the formation of
      hydroxyls. Some of the water reacts with the zinc that liberates a small portion
      of the  hydrogen ion from water molecules, causing hydroxyls to form (Zn +
      2H2O = Zn(OH)2 + H2).  WET believes that these "OH" radicals, along  with
      redox  shock, interfere with the normal cellular activity of bacteria and algae,
      thus reducing bio-film formation. Redox shock is the change in electrical (Eh)
      potential. As a rule, different types of bacteria and algae can only grow within a
      particular range of redox potential.  The passage of water through KDF causes
      a rapid and reversible reduction in Eh value of about 500 millivolts. This redox
      shift results in a disruption of electron transport and possibly causes a cascade
      of subsequent damage to the cell walls of single-cell organisms.

5.    Corrosion inhibition is achieved by the less-nobel metal, zinc, "sacrificing itself"
      to protect the other metals. In the case of KDF,  the zinc ions in solution are
      available to react chemically or  to sacrifice themselves to protect other metals in
      contact with  the water.

Outline of Process/Product

       In the  United States, over 700,000 cooling towers use water to  cool buildings,
process equipment, or products. Water is pumped through  a heat exchanger,
transferring the heat into the water. This water enters the top of the cooling tower,
passing into  the fill  areas and through forced air that evaporates  some of the water.
The evaporating water cools the remaining water  (by approximately 8 to 15 degrees
fahrenheit), which then re-enters the heat exchanger.


                                       167

-------
       If scale or algae forms in a cooling tower, efficiency is reduced, and dangerous
 bacteria (e.g., legionella) may develop. Also, the dissolved salts that remain when
 pure water evaporates can form scale that interferes with the transfer of heat  High
 salt contents become corrosive to the metal pipes, condensers, and structure of the
 cooling tower, which can cost millions of dollars in failed equipment and repairs each
 year.

       A multitude of chemicals (e.g., chromates, phosphates, silicates, pesticides
 herbicides, biocides) have been developed to treat cooling tower maintenance   '
 problems. During normal operation of the towers, these chemicals can enter water
 ecosystems and water treatment plants.  Many  chemicals have already been banned
 due to their toxicity, but some chemicals must still be used to address these
 problems.

       WET tested the KDF wool media as a replacement for chemicals to manage
 algae, bacteria, scale, and corrosion in cooling towers.  The test site was a comfort
 cooling tower in Fort Lauderdale, Florida.  Part of the water flowing through the tower
 system passed through floating modules contained KDF wool that were placed in the
 basin. The redox reaction occurred on a continuous basis during the daily operation
 of the tower. The flow through the KDF was approximately 6 percent  of the total tower
 flow per minute.

 Unique Product Features/Advantages

       KDF wool provides a natural control mechanism that raises the pH of the water
 which then reduces redox activity in a self-regulating mode.

       KDF is a unique, but common alloy, and once spent, the material mav be
 recycled through local scrap dealers.

 Process Schematic

       Cooling towers can vary in size from 8 to over 10,000 tons  in capacity  Water
flow rates can range from 35,000 to 43,000,000 gallons per day.

      A typical multi-chemical treatment (MCT) cooling tower is shown in  Figure 1
Water enters the cooling tower on  demand through the make-up water meter  A
circulating pump pushes the tower water through a heat exchanger and into the
cooling tower section where the water releases the heat acquired in the heat
exchanger. Chemical dosing pumps feed various chemicals into the tower basin
while a controller measures the salt content of the water via a conductivity cell  When
the salt content reaches a certain level due to  evaporation, the bleed manifold valve
automatically opens, and the water and chemicals are discharged  to drain
                                     168

-------
Figure 1.     Chemically Treated Cooling Tower
                                                                        REGULATING
       A KDF wool system (KWS) cooling tower is represented in Figures 2 and 3.
 Make-up water enters the cooling tower via a float valve. The cooling tower circulation
 pump sends water through the heat exchanger and back to the tower for cooling. A
 Total Dissolved Solids (TDS) controller measures salt content through a conductivity
 cell and a solenoid valve opens to release non-chemically treated water to drain. A
 separate low-volume pump recirculates cooling tower water through a prefilter to
 remove solids, then through a KDF wool contact chamber, and into the basin of the
 cooling tower.  Chemical dosing pumps are  not used.

        Initially KDF wool was placed in floating modules that had slots to allow water
 to flow across the wool.  The modules were placed in the cooling tower basin.  Water
 either fell into them directly from the fill area, or water was pumped into the modules
 so that water was recirculated on a 24-hour  basis.  When sediments  began filling and
 sinking the modules, a second generation containment vessel was developed.  The
 new "contact chamber"  was located next to and outside the cooling tower basin.
 Water  was filtered and then pumped into the chamber.  The water reacted with the
 KDF wool media and flowed back into the cooling tower basin.
                                      169

-------
Figure 2.     KDF Treatment Flow Diagram
             PREFtUTER
Figure 3.    Typical KDF System Layout
                                                                                REGULATING
                                     170

-------
APPLICATION

Process/Product Replaced

      KDF wool replaces traditional cooling tower chemicals. According to the March
1993 issue of Industrial Water Treatment magazine, "Over $750 million dollars worth of
cooling tower chemicals are used to treat approximately 500,000 'comfort  cooling
towers in the U.S. These towers are used for cooling hotels, resorts, and offices.
These figures do not include the tens of thousands of industrial and utility cooling
towers that use billions of gallons of water each year."  Some of the chemicals that
could be replaced include:

      Organophosphorus compounds
      Polyacrylate compounds
      Polymethacrylate polymers
      Chromates
      Polyphosphates
      Nitriles
      Silicates
      Heterocyclic compounds (mercaptobenzothiazole, benzotniazole)
      Chlorine
      Organosulphur compounds
      BJsthiocyanates
      Organobromines

 Wastes Prevented

      Treatment chemicals would be reduced or eliminated with the use of KDF wool.

       The number of labor hours needed to monitor and maintain cooling towers
 could be halved by using the self-regulating KDF wool media.  As KDF raises water
 pH  it slows its own activity.  As the lower pH make-up water enters the cooling tower
 basin, increased activity of the KDF raises the pH and once again stabilizes the KDF
 process.

       Water use is also reduced.  Chemically treated towers may drain constantly to
 prevent scale and algae build-up.  Since KDF wool is self-regulating, discharge rates
 may be significantly  reduced. Also, tests have shown that KDF towers can operate at
 higher salt concentrations than chemically treated towers, which results is less water
 usage.

 Cross Segment Uses

       KDF wool could be used in a variety of water treatment  applications:

       Cisterns or water storage tanks -- KDF wool can be used in  drinking water
       systems to reduce bacteria. For example, many countries use roof-top tanks to
       store drinking water for a building's occupants. These tanks become havens
       for biological  activity. EPA permits KDF to claim bacteria-static properties and
       has classified KDF as a pesticidal device, not a pesticide.
                                       171

-------
        im™           , use water for a variety of applications, such as water
        immersion therapy for burn patients. A patient's potential for bacterial infection
        is enhanced when the skin's protective barrier is removed  KDF wool treated
        water could eliminate harmful bacteria from the water prior to pattent trelfment.

        Hot water systems - Many hot water systems harbor bacteria, such as

        K±n±'Chh %m-e 15 If degree fahrenheit water.  Laboratory tette found
        through °he KDF                exposed to water that has been passed
        ^% m*etal reductions " Heavy metals are often difficult to remove during the
        wastewater treatment process.  KDF can remove part per billion levete of certain

        bMS^nSM *e NeW JerS6y DePartment °f Health showed ?hat KDF
        is the best technology for removing mercury from water.


        Chlorine removal -- Many wastewater treatment systems contain chlorine that

        rahna^en,penSIV? t0 Jiemove- Tests on a secondary waste™ stream in |t
        Charles, Ilinois, found that 57 pounds of KDF processed over 1  million gallons
        of effluent while removing all chlorine at a flow rate of 5 gallons per minute
 PROCEDURE


 DEMONSTRATION
  •   ^ Jwo exi!tin9 circular, fiberglass Protec comfort cooling towers located at a mw
 sized hotel in Fort Lauderdale, Florida, were selected for the projert


 ann th JhS C«°,°lin9 -ower 5n the 9raund floor 'evel of the hotel is rated at 700 tons
 and the roof tower is rated at 225 tons.  Both towers were being chemicallv treated hv
 SJS22? contract°.r- For.lhe trial- ^e ground-level tower con tinued to^bf cheSv V
 treated for comparison with the KDF wool-treated tower on the roof  WET etected n

            tOWer ^ ^ KDF tre                                           *°

                        KDF treatment since il offered       security


                    Z







      Special biofilm coupons, developed by Dr. John Wireman of Biolonicai
Solutions, were also installed to monitor the growth of bio films ^ The counnnQ
                                                    ms    e coun
made from a material that allowed biofilms toattach to the coupon surface
Measurements between the two towers could be compared        surrace
                                     172

-------
      For twelve months, WET technicians conducted site visits every two weeks to
monitor water chemistry - hardness, IDS, copper, zinc, iron, alkalinity, temperature,
and pH - and note conditions of the towers.  The corrosion coupons were replaced
every two months; used coupons were sent to an independent laboratory for analysis.
Biofilm coupons were collected on a monthly  basis and sent to Dr. Wireman for
analysis.

      During the last eight months of the project, other KDF wool cooling tower sites
were also being monitored.  At these sites, WET charged the owners the same fee as
they were paying for chemical treatment  (as EPA funding did not cover these sites)
and guaranteed equal or better results.  WET was concerned that, due to a vandalism
incident of the test roof-top tower, and damage to the tower inflicted by Hurricane
Andrew, they needed additional data to better understand the results obtained from
the Grant Program test site.

Evaluation Parameters

      The KDF and chemically treated cooling towers were compared for corrosion
rates, bacterial growth, and scaling tendencies.

Assessment Summary

      KDF wool controls scale and biofilm, and to a lesser  degree, corrosion.  A
vandalism incident and damage inflicted  by Hurricane Andrew did not permit a 12-
month uninterrupted test cycle at the EPA-funded test site.  However,  from information
collected at the 12 other test sites, WET  concluded that the average performance life
of KDF wool on South Florida water is 6 months.

Cost/Payback of Demonstration

      The final project cost was $42,297, with $24,830 contributed by EPA through
the Pollution Prevention By and For Small Business  grant program; WET contributed
the balance of $17,417.

      The  cost to chemically treat the test tower would have been approximately
$1,800. The cost to use KDF on a normal  basis would be comparable.

      Although WET had to replace the KDF media 3 times during the test project
due to various upsets, this led to the development of a second generation
containment vessel to hold the KDF media. In addition, WET learned  to filter the water
to remove large particles of scale and dirt before passing the water through the
media.

      WET can offer the KDF technology at  a similar price to chemical treatment, but
with less environmental impact, less water  usage, and reduced labor.


RESULTS AND DISCUSSION

       The  purpose of the pilot study was to  determine the  efficacy of KDF wool
process  media in lieu of traditional multi-chemical methods  in the treatment of


                                     173

-------
  problems common to water-based heat exchange systems.  During the study data
  were collected in the following areas:                           ay, ^*

        Corrosion rates
  •      Bacteria
        Langelier Scaling Index values
        Cycles of Concentration
  •      General water chemistry

        The pilot also allowed an evaluation of media application to similar systems.

  <0no *?ev®*ral Problems developed during the course of the pilot study.  In late May
  1992, the site was subjected to tampering by unknown parties.  Concurrently a
  recurring mechanical problem  with the tower's rotating distribution arm developed  In
  late August 1992, Hurricane Andrew passed through south Florida, damaging the site
  and causing system upsets. In the third quarter of the pilot (late 1992), the distribution
  arm malfunction was causing a serious problem, and termination of the study was
  considered if repairs were not  made by the tower owner.  The tower distribution arms
 were repaired in mid-January 1993.                                   IUUUUM *mib

       A cooling tower (not in the pilot study) was retrofitted in August  1992 with a
 new method of media application that utilized a newly developed contact chamber and
 a side-stream approach; this yielded better results than using KDF in the floating

 cmo°^                            WET C0uld not reP|ace ndS             3nd KDF Whi'e 3 fl°W reStriCt°r maint^ned  optaSm fbw     9
                                     174

-------
CONDITIONS THAT IMPACT PERFORMANCE

EH

      Redox reactions were optimized when water pH was below 8.5. When the KDF
wool reacts with the water, pH was raised and the KDF process slowed. The addition
of lower pH feed or make-up water reactivated the  KDF reaction.

Air

      Once the KDF wool is wet, exposure to air will shorten its performance life.
When the test tower was vandalized and drained, and when the tower was damaged
by Hurricane Andrew, the wet  KDF wool was exposed to air and dried. The wool
never regained optimum performance and had to be replaced.

Sediment

      A build-up of sediment in the KDF media only traps more sediment, and the
KDF becomes a filter rather than a redox  reactor.

Hydraulics

      A steady  even flow of water over the entire  KDF media surface is necessary to
optimize performance.  An  ideal flow rate is 10 gallons per minute for 5 pounds of
KDF wool.

TABULATION OF DATA

 LSI Values and Cycles

      This data illustrate the relationship  of the LSI and the Cycles of Concentration.
The LSI is commonly used as a  point of reference relating to the aggressive (or
 scaling) nature of the system water.  A range  of ± 1.0 indicates a relatively neutral
 water riot necessarily aggressive or scaling.  Cycles of Concentration is a relative term
 referring to the mineral concentration(s) of the system water versus the source water.
 The LSI and Cycles of Concentration are important, as they provide  a picture of the
 system that cannot be directly observed or quantified through other  means. With
 Cycles  of Concentration, as "mineral levels left behind by the evaporation of pure
 water" increase, the aggressive nature  of the system water may be minimized, and
 scaling tendencies become pronounced. However, high levels of chlorides will add to
 the aggressive nature of the system water and contribute to corrosion rates.  The
 Cycles of Concentration also indicate the number  of times the system water is reused
 before  it is discharged to drain.  Thus, the higher the cycles, the more efficient the  use
 of water in the tower.

       During the pilot period, WET was not aware of any scaling in  the heat exchange
 areas  The average Cycles of Concentration  for the KWS tower was 4.8353, and the
 average LSI value was 1.1106. The average cycles for the  MCT tower was 4.62, with
 an average LSI  value of 1.1677. The performance of the KWS tower was comparable
 to the performance of the  MCT tower.  A comparison of  LSI values and cycles in the
 KWS tower and the MCT (control) tower are shown in Figures 4 and 5 and in Table 1.


                                      175

-------
 Figure 4.     Pilot Tower - KWS:  LSI Values and Cycles
                              ''31/92 I 6/01/92 ' V3C-S2 I 9 /' • •'-? '' •-
                                 V29/92  7/CVS2  8/3 1/92" ':"1/2, 5";
                                                DAY or

                                                 LSi VALUES

                                                 AVE. CYCLES
Figure 5.     Control Tower - MCT:  LSI Values and Cycles
                                            LSI VALUES & CYCLES
                                             DAY or SAMPLE
                                               LSI VALUE
                                             *  CYCLES
                                           176

-------
TABLE 1.    KWS (Pilot) and MCT (Control) Towers:  LSI Values amd Cycles
KWS TOWER
Sample
Date
3/31/92
4/29/92
6/1/92
7/1/92
7/30/92
8/31/92
9/31/92
11/2/92
12/2/92
12/30/92
2/1/93
3/2/93
3/31/93
°C
24.4
25.8
27.9
28.0
30.5
28.1
26.3
27.6
28.5
25.4
19.8
24.8
27.2
LSI
Value
0.7699
1.1677
2.2842
2.0910
1.3217
1.0326
1.3677
0.6853
0.6639
0.7007
0.5270
1.1319
0.6945
Total
Averag
e
Cycles
3.20
8.44
6.79
8.24
2.69
3.61
4.49
5.82
3.13
3.58
5.21
3.95
3.71
MCT TOWER
Sample
Date
3/27/92
4/29/92
6/1/92
7/1/92
7/30/92
8/31/92
9/30/92
11/2/92
12/2/92
12/30/9
2
2/1/93
3/2/93
3/31/93
°C
27.0
25.8
27.1
27.8
28.7
27.9
26.7
27.1
27.1
26.3
23.6
24.1
27.9
LSI
Value
0.1535
0.8846
1.4225
1.4458
1.6165
1.0553
2.1977
0.6141
1.0580

0.4847
1.4463
1.6528
Total
Averag
e
Cycles
2.16
2.82
3.07
3.45
3.33
3.88
10.72
4.72
3.69
2.72
4.94
4.53
10.14
 Bacterial Data

      The relationship between the total bacterial plate counts in the KWS tower and
 the MCT tower is shown in Figure 6 and in Table 2. Initially, the KWS tower yielded
 much lower total plate counts than the MCT tower.  With a system upset due to
 tampering in  May 1992, bacterial counts were still lower than those seen in the MCT
 tower. However, counts spiked during August to September 1992, and KWS bacteria
 counts exceeded MCT counts.

      The first indications of system water distribution problems in the test tower
 surfaced in mid-July 1992. Since the distribution arms were not rotating, a full flow of
 system water through the KDF media was not possible. Bacterial counts began to
 rise.  In late August 1992, Hurricane Andrew caused a massive system failure, and the
 tower was pumped dry on several occasions, exposing the wool and wetted surfaces
 to ambient air.  The tower was also inoperable for several days with no flows through
 the wool media. When stabilized and  operational, WET noticed a significant drop  in
 total bacterial counts from the September 30, 1992 levels.  However, from that point,
                                      177

-------
      The annual average of the total bacterial counts were within 9 percent of values
from the test and control towers.  A significant deviation in bacterial counts in the KWS
tower was noted during the August/September 1992 period.  If this count of 2 8E6
were at a level of 1.5E6, consistent with the MCT tower, the annual average figure for
the KWS tower would have been approximately 666,182 colony forming units (CPU)
per month or 9 percent less than the annual average for the MCT tower.  From the
data presented, the total bacterial counts reported in the KWS tower were well within
the ranges of those seen in the MCT tower.


Figure 6.     KWS and MCT Bacterial Plate Counts
     II
     O =
                                     BACTERIAL PLATE COUN1
                  04/29/92   07/01/92 I  08/31/92   11/02/92   12/30/92   03/02/93
                       06/01/92   07/30/92   09/30/92   12/02/92   02/01/93  7

                                        DAY OF SAMPLE
                                       D   MCT CFU

                                       +   KWS CFU
                                     178

-------
TABLE 2.    KWS and MCT Bacterial Plate Counts
DATE
4/29/92
6/1/92
7/1/92
7/30/92
8/31/92
9/30/92
11/2/92
12/2/92
12/30/92
2/1/93
3/2/93
MCT TOTAL BACTERIA
2,000,000
2,000,000
900,000
740,000
240,000
1,500,000
240,000
16,000
260,000
36,000
33,000
KWS TOTAL BACTERIA
2,100,000
670,000
85,000
690,000
410,000
2,800,000
680,000
410,000
600,000
13,000
170,000
Corrosion Rates - Alloy

      Corrosion rate data compare the corrosion rates in the KWS and the MCT
towers on alloy metals. These metals,  due to their composition, are less susceptible
to corrosion in most environments and are representative not only of the metals in this
system, but of HVAC systems in general.  Upon initial observation, the corrosion rates
in the MCT tower are subject to radical fluctuations.  This may be due, in part, to the
adjustment of chemical feeds for pH and inhibitor control in response to the changing
nature of the source water chemistry.  The self-regulating nature of the  KDF, in
response to pH changes in the system/source water, allows for a more consistent
level of corrosion inhibition. KWS tower base-line corrosion figures gathered prior to
the initiation of the pilot indicate a rate  of 0.31 mils per year for rolled copper using
CDA 122 coupons.  The average annual corrosion rate for the same coupon during
the test with KDF wool was 0.431 mils  per year; the annual MCT tower corrosion rate
for copper was 0.530  mils.

       Base-line data were not collected for admiralty brass or stainless steel coupons.
During the test period in the MCT tower, the average annual corrosion rate for
admiralty brass was 0.677 mils, and  0.241 mils for 316 stainless steel. In the KWS
tower, corrosion rates were 0.798 mils  for admiralty brass and 0.249 mils for stainless
steel. The KWS tower performed favorable as compared to the MCT tower.
Corrosion rates in the KWS and MCT towers are shown  in Figures 7  and 8 and in
Table 3.
                                      179

-------
 Figure 7.      Pilot Tower - KWS:  Corrosion Rates
                                               CORROSION RATES
                            05-92   I  07-92  I   09-92
                                C6-92     08-92
 '   01-93   '
2-92     02-9:
                   l5 -92
                -92
             DATES
      0  ADMIRALTY BRASS
STAINLESS STEEL      O  ROLLED COPPER
J3-93
Figure 8.     Control Tower - MCT:  Corrosion Rates
                                             CORROSION RATES
                      1.1

                       1

                      0.9

                      0.8

                      0.7

                      0.6

                      0.5

                      0.4

                      0.3

                      0.2

                      0.1

                       0
                         05-92  I   07-92   I   09-92  I   11-92   i  01-93  I   03-93
                              06-92     08-92     10-92     12-92     02-93
                                                 DATES
                                         O  ADMIRALTY BRASS
                                +  STAINLESS STEEL     O   ROLLED COPPER
                                               180

-------
TABLE 3.    KWS and MCT Towers:  Alloy Corrosion Rates
KWS Alloy Corrosion Rates
Date
5/92
6/92
7/92
8/92
9/92
10/92
11/92
12/92
1/93
2/93
3/93
Adm.
Brass
(mils)
0.46000
0.79000
0.50000
1.13900
0.74260
1.48840
1.03750
0.91330
0.58090
0.55530
0.56810
Stain.
Steel
(mils)
0.15000
0.30000
0.19000
0.38600
0.10380
0.45910
0.52700
0.30930
0.11930
0.16320
0.02850
Rolled
Copper
(mils)
0.32000
0.62000
0.50000
0.54570
0.34260
0.43670
0.63060
0.52830
0.27140
0.25940
0.28260
MCT Alloy Corrosion Rates
Adm.
Brass
(mils)
0.47000
0.70000
0.62000
0.99530
0.57030
0.90000
1.03260
0.67930
0.73080
0.45750
0.29370
Stain.
Steel
(milsL
0.15000
0.30000
0.18000
0.46280
0.11610
0.49450
0.43980
0.27570
0.09360
0.07170
0.06540
Rolled
Copper
(mils)
0.32000
0.49000
0.34000
0.90870
0.15780
0.58940
0.86810
0.71350
0.22350
0.36960
0.85030
      Corrosion rate tests (measured in mils per year) for the KDF tower indicate
similar results for admiralty brass, stainless steel, and rolled copper, but the KDF tower
showed less variation in the acceleration of corrosion rates. WET feels this is due to
the constant adjusting of chemical dosing to match the ever changing feed water
supply (in chemically treated towers).

Corrosion Rates - Steel

      The chemically treated towers in the pilot study have lower corrosion rates for
mild steel and galvanized steel.  The data for steel corrosion rates indicate the
relationships between the KWS and MCT towers on untreated, C1010 mild  steel
coupons. Black iron coupons may be considered more  representative of system
metallurgy; however, the difficulties of correctly determining  corrosion rates on this
type of coupon  outweigh the prospective benefits of this data. The HD galvanized
coupons are representative of system metallurgy.

      Base-line corrosion rate data were collected from the KWS tower under
chemical treatment conditions prior to the initiation of the pilot study on both mild steel
(7.05 mils) and HD galvanized (4.95 mils) coupons.  During  the pilot test, the average
annual corrosion rates in the MCT tower were 9.225 mils on the mild steel  coupons
and 4.250 mills on the HD galvanized coupons.  When treated with KDF wool, annual
corrosion rates in the KWS tower were 16.418 mils for mild steel coupons and 5.889
mils for HD galvanized coupons.
                                      181

-------
Concentration with a similar source water chemistry. The results of the test indicated
an HD galvanized corrosion rate of 3.820 mils per year.  This is not conclusive yet it
may be presumed that at higher Cycles of Concentration, the corrosion protection of
milder metals would be better.  This is contrary to what one would normally expect to
see under such conditions.
      A comparison of corrosion rates in the KWS and MCT towers may be seen in
Figures 9 and 10 and in Table 4.
Figure 9.    Pilot Tower - KWS:  Steel Corrosion Rates
                                       CORROSION RATES
              24
              23
              22
              21
              20
              19
              18
              17
              16
              15
              14
              13
              12
              1 1
              10
               9
               8
               7
               6
               5
               4 -
               3
                 05-92
                       06-92
                            07-92
                                 08-92
                                      09-92
                                     D
   10-92
  DATES
MILD STEEL
                                                1 1-92
                                                     12-92
                                                          01-93
                                                                02-93
                                                                     03-93
                                  +   GALVANIZED STEEL
                                     182

-------
Figure 10.    Control Tower - MCT:  Steel Corrosion Rates
                                     CORROSION
                     05-92  !  07-92  I  C9-92  i  11-92  I  01-93  I  03-93
                         06-92    C8-92    '0-92    12-92    02-93
                                     2  ViLD S"£'_

                                   - -O. GAJ.AMZI
TABLE 4.    KWS and MCT Towers:  Steel Corrosion Rates
KWS Steel Corrosion Rates
Date
5/92
6/92
7/92
8/92
9/92
10/92
11/92
12/92
1/93
2/93
3/93
C1010 Steel
(mils)
11.71000
14.70000
11.92000
17.61000
17.43180
16.71370
22.40580
19.09190
16.52610
19.12480
13.36260
HD Galvanized
(mils)
5.54000
6.75000
6.48000
5.45500
6.40770
6.04510
5.14750
9.00530
4.00400
5.55950
4.38450
MCT Steel Corrosion Rates
C1010 Steel
(mils)
8.45000
6.39000
10.52000
11.26400
7.48710
12.51040
2.37490
10.17310
11.27130
9.91520
11.14290
HO Galvanized
(mils)
3.79000
4.47000
4.49000
4.23800
4.17340
4.63290
4.91410
6.64450
2.79700
3.85400
2.7415
                                         183

-------
 Biofilms

       During the first two months of the test, the KDF wool tower showed a
 remarkable ability to prevent the growth of biofilms as compared to the chemically
 treated tower.

       At the end of the second month, vandals drained the cooling tower, exposing
 the KDF wool media to the air.  From that point, biofilm tests did not show reduction
 levels as they had previously.  After 4.5 months, the KDF wool was replaced, as
 results from the vandalized tower were still substandard as compared to the other
 KDF towers.  In August 1992, Hurricane Andrew damaged the tower, causing the
 water to drain and exposing the KDF wool to the air. Again, the KDF wool was
 replaced.

       Recurring mechanical problems with the water distribution arms on top of the
 tower was causing uneven flows through the floating KDF modules. The KDF wool
 was replaced in October 1992,  and  by December 1992, biofilm tests still had not
 improved as expected.  In January 1993, building site management repaired the
 malfunctioning water distribution arms.  Fungi counts in the KDF tower then dropped
 to half the level of the chemically treated tower, although algae counts were higher.

       Given the  results of treatment generated through the use of the contact
 chamber design  on other sites of this nature, WET made a decision to discontinue the
 use of the floating modules in this application.

 Scale Control

      Once the KDF wool was  installed in the test tower, the amount of scale that
 broke free from the interior of the fill area and distribution system was high.  The
 debris sank several floating modules. Water hardness tests indicated that scale did
 not seem to form, even when the level of water hardness reached normal saturation or
 precipitation levels.

 COST/BENEFIT ANALYSIS

      Costs to treat a cooling tower with  KDF and with chemicals are virtually the
same, and the owners/operators of cooling towers would generally accept this new
non-chemical approach at the same value level  they pay for chemical treatment
services.  For the test tower, annual  treatment costs totaled $1,800 for each type of
treatment.

      Certain benefits for the cooling tower owner may be derived from using KDF
media:

      No chemicals stored on-site
      No chemicals discharged into the environment
      Better scale control, which reduces energy use
      Enhanced  pro-environmental image
      Reduced liability risks for chemical management

      Benefits for the cooling tower  service company include:


                                     184

-------
      No chemicals to mix or transport
      No chemical metering pumps to fail or malfunction
      More consistent treatment for customers
      Reduced labor costs by a contracting company
      Ability to recycle the spent KDF material
      Comparable profit structure
      Ability to increase territory and client base since fewer labor-hours are required
      to monitor and control KDF technology
•     Reduced occupational hazards for employees (as compared to chemical
      treatment methods)


CONCLUSIONS

POLLUTION  PREVENTION ASSESSMENT

Incentives

      The KDF technology is already patented.  However, in some countries that do
not honor USA patents or  even their own native  patents, the technology could be
copied.

      Advanced Water Management Group, a division of WET, has been established
to expand the technology in the Florida cooling tower market. Future activities include
introducing the KDF technology to comfort tower service companies who currently use
chemical treatment systems.  Advanced Water Management Group also hopes to gain
the interest of tower manufacturers and chemical producers who serve the cooling
tower industry.

Barriers

      Competitive challenges of the marketplace and cash flow are normal barriers to
any technology. WET needs additional funding to carry the KDF technology into the
larger industrial and utility cooling tower market.  WET must also educate cooling
tower operators and service companies of the merits of this new, non-traditional
approach to maintaining cooling towers.
                                    185

-------
                            PRE-CHARGED VACUUM
                LIQUID EXTRACTOR / CONTAINERIZATION DEVICE

                                      by

                                Lowell Goodman
                           Technical Support Services
                        SOUTH THOMASTON, ME 04858
ABSTRACT
Technical Support Services has developed and tested an innovative device that
provides simple, spill-free extraction of used oil from equipment during routine
maintenance activities such as engine oil changes.  The objective of this project is to
design, test, and evaluate this device for use in other applications. Successful
development of this device will  provide a mechanism to remove and contain materials
in otherwise diffucult extraction situations, thereby promoting recovery and recycling of
material which is frequently disposed.

INTRODUCTION

PROJECT DESCRIPTION

The Slurper™ is an  innovative, self-contained, pre-charged vacuum device which
provides simple, spill-free extraction of fluids - primarily used oil from marine engines
and industrial equipment during routine  oil changes.  Liquid is "vacuumed" into a
sealed, steel tank for spill-proof containerization until the  material is reclaimed.  The
objective of the project is to design, test, and evaluate modifications of this device for
other applications such as recovery of chlorofluorocarbons (CFCs), extraction of
automotive fluids from vehicles  at an accident scene or salvage yard, and extraction
and containment of bodily fluids by embalmers. Removing and containing hazardous
liquids in difficult extraction situations can promote recovery and recycling of materials
which may ordinarily be disposed. The  project included  design work, operational and
flow rate testing, and analytical  work and engineering studies.

Two single compartment models were designed.  A 4.75 gallon model serves those
uses which require an extraction capacity of less than 18 quarts (such as automotive
engines).  A 9  gallon design accommodates larger capacities,  such as diesel engines
which typically have a 5 gallon oil reservoir. The latter unit is targeted for commercial
applications. Both  units are configured with standard 3/4 inch drain valves for easy
draining and cleaning.

A •two-compartment unit was  also designed to remove waste oil by vacuum and
replace fresh oil by pressure.  (This concept evolved prior to receipt of the Grant and
was submitted with the original  Slurper™  patent application.) This unit has two sides
that perform the following functions:

•     "Take in" fresh oil by vacuum
•     Remove waste oil  by vacuum
•     Replace fresh oil by pressure
                                     186

-------
In the two compartment model, both halves are precharged with vacuum.  A suction
tube is connected to what is normally the "pressure side" of the vessel, allowing fresh
oil to be "taken in" by vacuum.  Next, air pressure is applied to the "pressure side" on
top of the fresh oil.  The unit is now ready to draw in waste oil via the vacuum side
and replace fresh oil via the pressure side.

Unique Product Features

The vacuum process can simultaneously remove and containerize a material, thereby
immediately preparing it for recovery.  Key advantages include:

      Application of a portable self-containerized vacuum
      Simultaneous containment of liquids or gases
      Safe in explosive  environments (no electricity)
      Application of a near absolute vacuum
      Chargeable by shop air pressure
       Extraction to  metal hand operated connectors (vacuum tight)
      97 percent fill capacity
       Light weight (no heavy pumps or batteries)
       Doesn't get "tired"

The last point is especially evident with highly viscous fluids.  The Slurper™ keeps
drawing on the material  with the same force until fill capacity is reached without regard
to time  required to extract the material. There is no additional vacuum  loss with a
thicker  liquid; vacuum loss  is dependent only on the amount of liquid actually taken in.
Figure 1 illustrates a one-compartment Slurper™.  Figure 2 shows the two-compartment
model.

Figure 1.    One-compartment Slurper™
                    The figure below outlines (he SKrper"* pert by pert. Number* have been •s*yi«d to
                    me representative pert.1


                       1  STEEL TANK (4.75 GALLON OR 9 0 GALLON)
                         AIR PRESSURE MPUT (TIRE VALVE FOR STANDARD AIR CHUCK)
                         TWO CTAOC VBNTURI VACUUM OCVICC
                         MWM-VACUUM OmCK VALVI
                         VACUUM OAOE («0* Of MERCURY)
                         TUBMQ TO METAL HAND OPERATED CONNECTORS (VACUUM TKJHT)
                         Cf*Off CONTROL VULVE
                         ALTERNATE HOSES (1/4* I.D. AND Vl«' IJJ)
                         DRAM VALVE (»O
                         MISCB.LANEOUS BRASS B.0OWS AND PTTTKJS
                       Q


\ I 	
^••'~ "^ •
H L .
^-^\
* — — — -
i



/IN
( f j




// £
it /
r1-1!
_L
                                  VACUUM
                                         187

-------
 Figure 2.     Two-compartment Slurper™
                            VACUUM
                             SIDE
                            WASTE
                            LIQUID
PRESSURE
  SIDE
 CLEAN
 LIQUID

APPLICATION

Process/Products Replaced

For marine oil changes, the Slurper™ replaces conventional manual and 12 volt electric
pumps mounted on a plastic bucket with an opening in the lid.  These existing
commercial pumps are low pressure fluid pumping devices and by design are
inherently messy because the oil actually goes "through" the pump itself. These
pumps function by pushing air forward which creates a "siphon" in the suction hose to
initiate the pumping process.  Once the oil  is pumped into the bucket, the 12 volt
motor must be reversed to pump the oil from the bucket; the oil can also be poured
through the opening in the lid.  Most pumps require close proximity to a battery or
electric source for power, thereby limiting their portability.  The electrically operated
units are potentially hazardous in the bilges of boats with gasoline engines.

Wastes Prevented

No-spill vacuum containerization of used oil can enhance its recyclability, as
containment,  storage, and transpprt is simplified.  The containment process eliminates
the introduction of oil into the environment, as the oil is transferred directly from the
engine into the Slurper™.  Storage and transportation have no  risks of spills.

With any liquid, if containment and storage  are time consuming, awkward, or
expensive, the individual or business may default to a more simple, but improper
disposal method.  For example, it is easier  to pour old oil from  a bucket into the
sewer, rather than risk spilling oil in the trunk of one's car during transport to the
proper recycling site.
                                      188

-------
Cross Segment Uses

Industries finding useful Slurper™ applications include: plumbing and heating
contractors, trucking and fleet companies, rental vehicle fleets, automotive repair
stations, and machinery in virtually any industry.

Innovative applications include:

•     Embalming - Embalmers do not always receive notification  of an HIV positive
      cadaver, and concern over safe management of bodily fluids is prevalent in the
      funeral industry.  Vacuum technology permits the evacuation and containment
      of bodily fluids during the embalming process.

•     Holding Tank Cleaner - Boats are not permitted to discharge the contents of
      the holding tank overboard.  The Slurper™, modified with a larger diameter
      hose, can easily evacuate the tank.

      Welding Solder Extractor - Excess solder can easily be removed from printed
      circuit boards and containerized for subsequent recycling.

      Oil Furnace  Priming - When the Slurper™  nozzle is attached to the bleed point
      on an oil furnace,  priming may be quickly accomplished  (especially during cold,
      winter months) when  compared to traditional gravity bleed  priming.


PROCEDURE

DEMONSTRATION

Test Conditions

Given the broad spectrum of materials which could be handled by the vacuum
technique, tests and evaluations were conducted with:

•     Liquid Containment
      —     Water (control  liquid)
      —     Gasoline
      —     Kerosene
      —     Linseed Oil
      -     Motor Oil (SAE 30)
      —     85/140 Gear Lube

•     Gaseous Containment
      -     Air
      —     CFCs (Freon)

The liquid and  gaseous tests encompassed percentage fill capacities and flow rates
(suction characteristics).  They were also designed to demonstrate operational
boundary conditions for liquid containment within the containment sizes available.
These basic designs were operationally tested to provide baseline information on
performance and potential.
                                     189

-------
 Flow rate tests were performed on a representative sample of materials.  These tests
 were structured and controlled to provide representative data and to describe
 relationships among the operational parameters which significantly affect each of the
 flow rates.

 Given the complexities of fluid dynamics, a thorough test series of all fluid dynamic
 factors was considered outside of the scope of the project.

 The tests and test results for liquid flow are not quantitative to the Bureau of
 Standard's laboratory conditions, particularly in temperature control.

 Boundary Conditions

 Several tests were designed to present low end boundary  conditions, where possible
 For example, 85/140 gear lube was tested at 0° F with a 7  foot long, 1/4 inch internal
 diameter (ID) tube.

 In designing boundary condition tests for technique, "everyday products" were tested
 to provide quick representation of low end response for highly  viscous materials.

 Repeatability was consistent throughout the liquid tests.  Test results variations were
 mostly attributed to accuracy of gauge readings, control value  (to stop watch time),
 and liquid test measurements, including estimating stopping points for "thinner" liquids.

 Multiple samples were taken for statistically valid sets.

 Two Compartment Tests

 Pressure tests were configured to present flow rates for 25 and 50 psi on  replacement
 liquids (water and  15/40 oil at room temperature).

 Pressurized gas flow tests were not performed due to difficulty  in  arranging controlled
 experiments reflecting gas replacement under pressure.

 Liquid Tests

 Liquid flow rate tests reflected various system  configurations including:

      Starting Vacuum - 5 inch increments except for tests of near absolute vacuum
      (either 28 inches or 29 inches used)

      Hose Diameter and Length - diameters  varied  between 3/16 inch and 1/4 inch
      ID; lengths up to 90 feet were tested

      Viscosity - all material viscosities varied  as a function of temperature

      Temperature - 30° - 180° F range tested; gasoline tested  at room temperature
      only

•     Height of Tank Placement - above or below liquid
                                      190

-------
Flow rate test results are presented for the most meaningful interpretation of data. For
example, oil tests showing flow rate data at 0° F are more difficult to interpret in terms
of 0.0036 quarts per minute rather than 13 minutes for one quart.

Gaseous Tests

Gaseous tests were conducted on air and Freon using the following parameters:
•     Starting vacuum
•     Hose diameter
•      Hose length
•      Height of tank placement - this parameter is not considered to impact this test
       due to low mass density of test gases

Outside the capabilities of test controls was qualitative information of Freon under
pressures sufficient to cause a phase change to a liquid state.

RESULTS AND DISCUSSION

PERFORMANCE RESULTS

Fill Capacity

Tests results show a marked difference between fill capacity characteristics of a near
absolute vacuum versus a partial vacuum for a given loss of vacuum.  Although the
amount of fill is proportional to starting vacuum, the amount for each inch (of mercury)
of fill is significantly different (a factor of 10).  Figure 3 illustrates the first inch fill
capacity of a 9 gallon unit.

Figure 3.    First Inch Fill Capacity:  9 Gallon Unit
                    6-
                    2-
                            NEAR ABSOLUTE     VS     PARTIAL VACUUM
                              FIRST INCH FILL. 90 GAL UNIT USING WATER
                                        TEMP 60* F
                      4.66

                          298
1
1
                                 178  ,43
                       29  28  27   26  25
                                STARTING VACUUM
                                (Inches of Mercury)
                         NEAR ABSOLUTE

                           VACUUM
                      |^UL&JU&JU£UU&iliJUUfJULa

                         FIRST INCH FILL
                         466 GALLONS
                                               0.49 0.47 o.43  0 42 0 40



                                                15  14   13  12  II
                       FIRST INCH FILL
                        0 49 GALLONS
                                         191

-------
Total fill is proportional to starting vacuum in most cases as seen in Figure 4.
Figure 4.    Comparison of Fill to Starting Vacuum
             8.67
                         NEAR ABSOLUTE      VS       PARTIAL VACUUM
                           TOTAL FILL FROM STARTING VACUUM TO ZERO
                                  9.0 GAL. UNIT USING WATER
      3
      ±  4
      Lu
      h-
      o







XV
SS^
\V
1
1
\V
v\>
^ X \
\v








ss.
XN
xx

xx
s\\
i
I








SS
NX

XX
^\
i
i







/./»•

SSV

XV
XX
E|
1
^







7.4t

1
XX
XX
1

*








4-


i
!::
i
i
jjH
•!?!
||
ilij
n



4t
:|!

|
jj!
i
n
m
if-



3






4 1C


•:;:•';:::
iiiiiiill
;:!::•:;!
hlijiJ!'
:ii!:j:i:
!£$:!
Ill
II
iljiiilil
!:!:!!!•:
I!!!!;!!!



>







3.3^

iii
;• ••••••i
'i !:!i;
!!:»;!:'
m
ij-jjiji
Hi




>







355
:||:j:-:-? *••••:
IliiPli
ttii tii-
11 ,:
II - i
             29   28   27   26   25
                y            STARTING VACUUM
                             (Inches of Mercury)
5   14   13   12
  v

                                                      \
                     \
                                                       \
                                      90 GAL. UNIT
                                                      TOTAL FILL
                                                     A 48 GALLONS
                                    192

-------
However, when operating with hot liquids, especially in the partial vacuum range, fills
for the first inch (of mercury) are significantly lower than those at room temperatures.
Both hot liquids and highly volatile liquids partially vaporize upon coming into contact
with the vacuum, creating an immediate vacuum drop of several inches (under partial
vacuum). For example, water  showed a reduction  of total fill from 4.48 gallons at 60°
F to 3.16 gallons at 180° F at 15 inches vacuum - a performance loss of 1.32 gallons.

At near absolute vacuum, hot liquid has a negligible effect on vacuum loss or capacity
during the first few inches of fill as the actual percentage of air in the tank was  almost
nil.

"Head losses" are directly proportional to the elevation difference between the liquid
and the vacuum as seen in Figure 5.
Figure 5.     "Head" Loss Results
                  STAR TING VACUUM
                    29 INCHES
                   9.0 GALLON UNIT
                  STARTING VACUUM
                    29 INCHES
                                     T
                                     I OFT.
                                       193

-------
Flow Rates

Figure 6 represents the baseline data set using water as the "control".  Flow rates,
even for water, show a perceptible change as a function of viscosity and temperature.
These flow rates incorporate the effects of temporary vacuum loss during the first few
seconds of flow due to the liquid-to-vapor phase change.
Figure 6.    Flow Rates: Water
               FLOW RATES
                  30.
                  25-
                  20-
               O v-
               < o
                  10-
FLOW RATE:

HOSE DIA.

HOSE LENGTH

TEMP

TEMP :
                                  345

                                   QUARTS PER MINUTE
Figure 7 illustrates the frictional effects of an extended length of suction hose on flow
rate.  At near absolute vacuums, the flow rate is approximately 6 times faster through
a short (7 feet) hose.
                                       194

-------
Figure 7.     Frictional Effects of Suction Hose on Flow Rate
                      30.
                      25.
                      20.

                     L\
                      10-
FlOW BATE

HOSED) A.:

TEMP.:
WATER

1/4 IN.

eo-F.
                                          HOSC LENGTH  90 FT.
                                    234

                                     QUARTS PER MNL/TE
In larger diameter tubing, the flow is turbulent.  Although extraction rates are very fast
water at 60° F: 1.6 gallons  per minute (gpm) for a 1/4 inch diameter hose vs. 28.6
gpm for a 3/4 inch hose -  a long, large diameter  suction hose will significantly reduce
the starting vacuum due to the air volume in the hose itself.

Flow rates for oil as a function of temperature and viscosity is seen in Figure 8:


Figure 8.     Oil Flow Rates as a Function of Temperature and Viscosity
                      30.
                        *
                      15- 11
                      10.li
                         I

                        'V
         OIL. 5.A.E. 30WT.

         1/4 IN.
                             to
                                   20     30     40
                                   MNUTE5 PER OUAR"
                                                     50
                                                           60
                                         195

-------
 Figure 9 shows the relative flow rate relationships among test materials of extreme
 viscosities at 60° F.
 Figure 9.    Flow Rates: All Test Materials
           E  7.
           ct.
           UJ
           Q.
                         FLOW RATE

                         HOSE DIA.

                         TEMP :
            ALL POLLUTANTS

            1/4 IN

            60* F
HOSE LENGTH :  7 /FT.

VACUUM      28 INCHES(HG)
                GEAR LUBE 30WT. OIL  LINSEED KEROSENE  WATER   GASOLINE
                                  OIL
Pressure Return Results: Two-Compartment Unit

When the two-compartment model is used to remove waste oil with a vacuum and
replace clean oil with pressure, one must start with adequate pressure to still have
useful pressure near the end of the operation. Since the replacement oil is most likelv
at room temperatures, it will follow the same flow rates under equivalent pressure as
represented by the previously shown vacuum flow rates.  As illustrated in Figure  10
the amount of useful pressure remaining is dependent on the volume of starting liguid
as well as the starting air pressure.                                           H
                                      196

-------
Figure 10.
      Useful Remaining Pressure as a Function of Starting Liquid
      Volume and Air Pressure
                               9 GALLON CONTAINERS
                                /       \
            25 P5I STARTING PRESS.
              STARTING W/8GAL.
25 PSI STARTING PRESS.
              3  P5I WhEN EMPTY
                                     25 PSI STARTING PRESS.
                                              :;x;xvJ  STARTING W/1 GAL.
                                                  25 PSI STARTING PRESS.
                                                   22 PSI WI-EN EMPTY
Worst Case Boundary Condition Tests

A variety of "worst case" or very thick liquids under very cold conditions were
evaluated; these tests were performed qualitatively without any attempts to represent
flow rate data.  (See Figure 11.)

Tests were run using granular materials (powders, flour, corn meal, laundry detergent,
etc.).  Extraction of all materials proved successful.

The above listed samples were then mixed with water to test thick slurry feasibility.
With the "pastier" mixtures (flour,  corn  meal), it was noted that the extraction hose
became slightly "lined" with the slurry.  If a higher proportion of water was added  to
the granular material, the hose was "washed down" with the vacuum, cleaning the
hose.

Different materials with other than smooth liquid textures have properties that require
specialized pick-up devices such as "sunflower heads".  Also, powder and granular
materials are not solid; their collection  is necessarily accomplished with reduced
volumetric efficiency.
                                        197

-------
Figure 11.
Worst Case Boundary Conditions
                                      WORST CASE
                                  BOUNDARY CONDITIONS
                                  NEAR ABSOLUTE VACUUM
                                   (28 IN. OF MERCURY
             WORST CASE - EXTREME COLD
                  OILS AND GREASE
                          WORST CASE - VISCOSITY EXTREMES
                               "THICK "AND GRANULAR
            65/140 GREASE   I QT. - 78MIN.
            _NO MEASUREABLE LOSS OF VACUUM

            15/40  OIL       1 QT. - 11 M~N~
            NO MEASUREABLE LOSS OF VACUUM
                                                      CONDITIONS
                                                     1/4 IN. I.D. TUBE
                                                      TEMP.  70' F.
                                SUGAR       I  QT. - 54 SEC.
                                 0.5 IN. LOSS OF VACUUM
                                SOAP
                                POWDER
1 QT.  -  47 SEC.
                                                       1  IN. LOSS OF VACUUM
                                  OTHER COMMON HIGH VISCOSITY
                             SUBSTANCES EXTRACTED FOR COMPARISON
                                  PEANUT
                                  BUTTER

                                  HONEY

                                  BUTTER
                               MAPLE SYRUP

                               APPLE JELLY

                               CORN MEAL
                                          198

-------
Qualitative Conclusions on Vacuum Performance

1.     The percentage fill of the container is directly proportional to the starting
      vacuum (for cold liquids).

2.     The percentage fill of one inch of vacuum -- 28-27 inches vs. 15-14 inches ~ is
      not the same.

3.     The physical height of a vacuum above a liquid causes an effective loss of
      approximately one inch of vacuum per foot of elevation.


4.     A liquid's viscosity has zero effect on the amount of vacuum used during the
      extraction process.

5     Hot engine oil (approximately 180° F) generally flows at approximately 1/2
      gallon per minute in 1/4 inch ID tube.  Oil at room temperature  (approximately
      65° F) slows to 1/2 gallon per 10 minutes using the same 1/4 inch ID tube.

6     Water at room temperature flows at approximately 1/2 gallon per minute
      through a 1/4 inch ID tube (roughly equal to hot engine oil).

7.     Flow  rate is approximately proportional to the square of the suction tube
      diameter.

8.     A container charged to 29 inches of vacuum will fill to 97 percent of tank
      capacity.

9.     A container charged to 15 inches of vacuum will fill to 51 percent of tank
      capacity.

 10.   The vacuum can only pull a liquid  (near the weight of water) to a maximum
      elevation of approximately 33 feet.


 Cost/Benefit Analysis

 The savings using the Slurper™ vacuum technology are not so much in dollars as they
 are in volumetric reduction of improperly disposed waste oil. In the automotive field,
 "do-it-yourselfers" improperly dispose of approximately 120 million gallons of waste oil
 annually.  If a Slurper™ rental/recycling process only saved 1  percent of this oil, 1.2
 million gallons of oil would be recovered.  Since one gallon of oil  can  contaminate one
 million gallons of drinking water, this 1 percent reduction could save 1.2 trillion gallons
 of water.

 Numerous non-monetary factors enter into the technology application:

 •     Spill  reduction from multiple transfers of waste oil.
      Increase in  proper management of recovered material.
 •     Production  and purchase cost savjngs for vacuum technology  as compared to
      currently available commercial devices.


                                       199

-------
  CONCLUSIONS


  POLLUTION PREVENTION ASSESSMENT

  Incentives
 During the project, response to the containerized vacuum technique was extremelv
 positive. Demonstrations and interfaces with businesses, private individuals
 municipalities, and rescue squads generated the most positive feedback on' the
 aspects of simplicity, containment, safety, and cleanliness of the operation.

 Phase of Development

 The technology is mature and in the marketplace.  The present configuration is
 suitable for small quantity liquid applications under 8.5 gallons.  Container size and
 M«a£e< ™n  n modlfl®d. to suit various applications.  Inquiries have been received for
 up to iuu gallon models.

Accessories were  developed to facilitate specific applications. One example is an
extraction hose coupling which is fitted directly to the oil dipstick tube on properlv
configured marine engines.                                          H-v^cny


Target Industry Potential


          tar&elrHarkfS,raun?|e trom "do-it-yourselfers" to small and medium sized
                                     The a^°ximate order of "pecw ^
                        ,u
           ^                          The a^°ximate order of "pecw
       Marine
       Industrial
       Automotive
       Other oil related extraction process
       Commercial rental
                                         configurations making it adaptable to

Commercial rental of the Slurper™ has strong potential.  In a "pilot" marine rental
program, pleasure boaters and commercial fishermen paid $10-$15 (depending on
                                        The unit fC" of * *as then
A two-compartment Slurper- could be rented as an "all-in-one" oil change kit  The
  *                                                        " and «lled
                                     200

-------
Barriers

Container size (currently nine gallons capacity is the largest) is a limitation.  There is a
practical limit to the size of the container which must be transported when full of liquid.
While the portable nature of the Slurper™ is a great convenience, some degree of
conservatism must be exercised to optimize the containerized vacuum.

Misunderstandings as to the nature and use of vacuum, especialjy containerized
vacuum, are common.


Potential Solutions

Understanding the engineering and non-engineering factors of vacuum techniques
can solve the above problems.  Availability of concise information and demonstration
of portable, containerized vacuum principles will allow the Slurper™ to gain widest
acceptance.
                                      201

-------
                 ENVIRONMENTALLY SAFE FOUNTAIN SOLUTIONS
                         FOR THE PRINTING INDUSTRIES

                                       by

                                 David R. Johns
                       Summit Resource Management, Inc.
                              Fort Wayne, IN 46804
 ABSTRACT

       Summit Resource Management, Inc. formulated an ecologically compatible
 fountain solution for the printing industry.  This new solution eliminated isopropyl
 alcohol and mineral acids. Ethylene glycol was also reduced through the substitution
 of propylene-based glycol.  These reductions and substitutions decreased the levels
 of harmful vapors in the air and reduced the amount of toxins released into the
 printer's wastewater system.


 INTRODUCTION

 PROJECT DESCRIPTION

       The off-set printing industry produces magazines, posters, brochures, cards,
 forms, and the like.  An ingredient in the fountain solution used  in the printing process
 is isopropyl alcohol (IPA), which is used to produce an etch that does not allow the
 transfer of ink to certain areas of the paper.  Printers typically use 5 to 25 percent IPA
 mixed in the fountain solution.  IPA generates harmful vapors, and some  amount is
 generally disposed through the wastewater treatment system.

       Summit Resource Management (Summit) formulated an alcohol-free,
 environmentally safe fountain solution. The definition "environmentally safe1' had to
 meet the wastewater disposal requirements and strive for a vapor pressure level of
 zero. One goal was to eliminate such ingredients as ethylene glycol, glycol ether  EB
 heavy metals, mineral acids, phosphates, alcohols, and low boiling solvents.

      After defining the need for certain properties in the traditional fountain solution
 Summit searched for environmentally friendly ingredients that would provide the same
 function. Such performance functions include surface tension, viscosity cleaning
 capability, compatibility with the inks, pH needs, and vapor and  odor concerns.


 Unique Product Features/Advantages

      The reformulated fountain solution is made with food grade products that are
 available from domestic manufacturers.

      Summit Resource Management developed two fountain solution formulas- one
for offset web presses with high speed printing, the other for sheet-fed presses that


                                     202

-------
print with slower speeds but higher quality.  Both formulas have the same ingredients,
but at different ratios to respond to different printing conditions. The products have a
pleasant odor, are environmentally safe, and perform well.  The products go into
solution easily, making it simple to mix and clean up.  Buffering agents stabilize the pH
differences found in municipal water supplies. A softening agent was also added to
balance the demands of hard and soft water; when using very hard water, all that is
required is a 1 to 2 ounce increase in product concentration.

Process Schematic

      Figure 1  shows a printing roller configuration and the pan where the fountain
solution is introduced on the press.   (Other variations include spraying the fountain
solution on the water form roller and brushing the water solution mixture over the
chrome roller.)  The  rollers generally are made from rubber, and the numbers on the
diagram indicate the softness,  or  durometer, or the roller. A low number indicates a
soft roller; a hard roller is numbered  over 30. A soft roller provides the best results
with the Summit product.

Figure 1.     Roller Configuration  and Fountain  Solution Pan


                  Roller Duroraeters
                    20 to 24
APPLICATION
 Process and Product Replaced

       Summit's alcohol-free product replaces traditional fountain solutions plus
 alcohol. The pressman only needs to work with one product instead of two.

       The required set-up process for using the Summit product is the opposite from
 the traditional process used by most press operators. When using the Summit
 solution, the pressman must start with  a clean press, maintaining that cleanliness
 throughout the printing process. The pressman must then turn off the ink dials and
 set the water dials so that water covers the rollers.  Once the rollers are clean the ink
 rollers are adjusted to the desired  ink density.  Traditional practices call for adjusting
 the water dials to the amount of ink.
                                       203

-------
 Wastes Prevented

       The Summit product allows the pressman to run a clean operation  Ink is
 reduced by 10 to 40 percent, and if set up correctly, the excess ink around the
 presses is eliminated.  The wastewater is also cleaner.

       Not only are environmental and health benefits realized through the elimination
 of alcohol process improvements can also be noted. Alcohol attacks the inks and
 actually reduces the effective level of the ink.  Alcohol eventually hardens or glazes the
 rubber rollers, preventing the transfer of solution to the rollers.  Once these  rollers are
 hard, they must be rebuilt or replaced.


 PROCEDURE

 DEMONSTRATION PROCEDURE

       Summit developed an initial alcohol-free formula and tested samples for volatile
 organic compounds (VOCs) and vapor density.  They then worked with several small
 printers to press-test the product.


 *  *  ^varly,res,ul? we^e mlxed'  ln most cases' the Product gave comparable results
 to traditional solution, but problems developed after a time:

       When the press  stopped, the printing plate would oxidize, which then had to be
       cleaned.

       The new solution was more sensitive to changes and would not compensate for
       the use of excess ink.
nn tho^oTTl'' readJus+ted the formulas by adding gum (thickener) to deposit a coating
on the plate to prevent oxidizing. The formula was also rebalanced to increase
cleaning.

      Once the formulation was refined, fifteen trial printing jobs were conducted at
various printers.  When conducting a trial, each pressman ran his own equipment
under normal conditions.  All colors were assessed from one-color to six-color
presses.
HM«  n  Su™mit foun.ta'n solution requires the rollers and blankets to be thoroughly
clean  Once the press is clean, 4 ounces of solution is added to a gallon of water and
placed m the water circulator.  The ink dial is set to the lowest level. The wate^r ro leT
1^11° C2mpjelely T**?6 .r°llers'  The ink rollers ^en are started and aojusted to
the desired ink density; the ink setting will be lower than when using traditional
Four press trials are described in Table 1.
                                     204

-------
TABLE 1.    Press Trial Parameters
PARAMETER
Equipment
Customer
Colors
Paper
Job
Concentration
of Summit
Product
Water pH
Water pH
After Addition
of Fountain
Solution
Roller
Durometer
Roller Wash
TRIAL 1
Two-color
Hiedelberg,
36-inch press;
currently
running 15%
IPA solution
Off-Set One,
Fort Wayne, IN
Red and black
Sixty pound
coated sheet
fed
5,000
brochures
4 ounces per
gallon of city
tap water
10.5
4.5
29 on ink roller
Mineral spirit-
based (not
water soluble)
TRIAL 2
Harris four-
color web,
slow speed
(600 feet per
minute)
James River,
Kendallville, IN
Red, yellow,
green, black
Boxboard
(cereal box
material)
Cereal boxes
3 ounces per
gallon of city
well water
9.0
4.6
25
Mineral spirit-
based
TRIAL 3
Kohl&
Madden sheet
fed, 36-inch
press, 2 color
Robert
Williams
Graphics, NY
Red, blue
Recycled
bond paper
Forms and
information
sheets
Started at 4
ounces per
gallon of city
water;
increased to 5
ounces after
the run was
started
7.8
4.7
25
Two-part,
water-based
TRIAL 4
Four-color
Komari
Container
Corporation,
Fort Wayne,
IN
Red, yellow,
blue, black
40 pound
coated
Outside
packaging for
baseball
cards
4 ounces per
gallon of
water ^
10
4.8
29
Aromatic 100-
base
petroleum
                                     205

-------
Procedure
















The press
operator
cleaned the
rollers and old
fountain
solution from
the equipment.
He followed
Summit's
instructions in
setting up the
press. The
pressman was
concerned
because the
ink dials were
set quite low.
The press
operator
cleaned and
prepared the
press
according to
Summit
procedures.









The pressman
cleaned the
old ink and
fountain
solution from
the press and
ran the
Summit
solution
according to
instructions.






Between
shifts, press
operators
quickly
cleaned out
old fountain
solutions and
wiped the ink
rollers. The
pressmen
manually
mixed the
Summit
solution and
followed
procedural
instructions.
EVALUATION PARAMETERS

      The parameters for evaluating the success of the Summit product were the
same as for current products: dry time, dot gain, press speed, toning, scumming,
number of impressions before cleaning or down-time problems, and slurring.  Summit
also focused on temperature and humidity levels, as changes in these two factors
affect plate oxidation levels and  ink acceptance.

      Figure 2 is a Graphic Arts Technical Foundation chart for normal and non-
acceptable printing that was used to judge the final printed product.
Figure 2.     Normal and Non-acceptable Printing Standards
            NORMAL
SIMULATED DOT GAIN   SIMULATED DOUBLE    SIMULATED SLUR
                                    206

-------
      Samples of the Summit product were tested by an independent laboratory for
VOCs arid vapor pressure.

      Summit established the following guidelines for pressmen using the Summit
solution, as the new procedures differed from traditional operating practices.

1.     Mix the Summit solution, 4 ounces per gallon of water.
2.     Clean the rollers and check their durometer (softness); softer is preferred.
3.     Set the ink  dial to the lowest markings. Set the water dial low,  but ensure the
      water roller is covered with the water/fountain solution mixture.
4.     Start the press and adjust the ink density to the desired levels.

      The success or failure of the print quality was based on the skill of each press
operator, who would evaluate his work using such questions as:

•     Were the dots that made up the actual print clean?
•     Were the edges sharp?
•     Were the colors the correct density and shade?
•     Diid the inks dry fast enough so as not to smear?
•     Is the printed material saleable?


ASSESSMENT SUMMARY

      Trial and error in formula development while working with the various pressmen
was critical to the success of the project. For example, Summit found that no two
presses performed alike, even if from the same manufacturer.  Press operators tend to
customi2ie their equipment to their liking.

COST OF DEMONSTRATION

      The total demonstration cost was $46,781, with $22,450 contributed  by EPA
through the Pollution Prevention By and For Small Business program, and the balance
provided by Summit.


RESULTS AND DISCUSSION

PERFORMANCE RESULTS

      Summit developed two products instead of the planned one, as a single
product could not handle different speeds of the presses. The basic ingredients of
the two products  are the same, but  are combined in different ratios.  Summit learned
that when the presses ran faster -- such as with web presses - smaller amounts of
the etch must remain on the rollers.  Summit needed to increase the cleaning
capability of the solution, leaving the correct viscosity of solution on the plate. The
concentration of the product simply  could not be diluted  because the viscosity of the
water on the rollers was not high enough to cover non-print areas.  In comparison,
alcohol is consistent in viscosity at a specific rate.  A ten  percent level of IPA  always
yields the same measurements, even when the fountain solution is dirty or weak in
etch concentrations.
                                     207

-------
       Summit altered the polymer level and increased solvency of their product to
 accommodate the differences of a clean fountain and a dirty one.

       Summit also found that measuring the concentration of the fountain solution
 was unrelated to solution conductivity.  Traditional fountain solutions contain small
 amounts of salts and metals that allow  a conductivity meter to measure levels of
 fountain solution in the pan. As the concentration of solution drops, a mixer
 automatically adds more fountain solution.  Summit solution additions are based on
 volume only, and Summit had to educated press operators that conductivity is not a
 key factor in the performance of a fountain solution.

 PRODUCT QUALITY VARIANCE

       Summit found little quality variance in the production of their new fountain
 solution. The project is a pleasant smelling and looking material, is easy to clean  and
 does not irritate the skin.  The product was designed as a concentrate, with a
 recommended dilution of 4 ounces per  gallon of water (compared to traditional
 product dilutions of 6 to 12 ounces per gallon of water).

 CONDITIONS THAT IMPACT PERFORMANCE

       Using the Summit product requires control changes in the ink and water levels
 as well as close monitoring of the press. The  Summit solution requires a consistently
 clean press.  When alcohol is used, the press operator does not have to be
 concerned with press cleanliness until the print job is finished.  Printers  must maintain
 a high volume operation, and Summit found that printers always were not willing to
 patiently fine tune the ink and water levels for the new solution or to clean the rollers
 sufficiently.

       High temperature and humidity were initially troublesome, creating oxidation of
 the plates, but the final product formulation solved the problem.  Conducting the press
 trials in the winter with low temperature  and humidity -- ideal printing conditions -
 made it challenging to judge the effects  of these two variables.

       Measuring the number of ounces of fountain solution was difficult for some
 larger printers that use conductivity meters to determine the proper time to add
 solution.  Their systems were automatic  and could not measure  volume  Automatic
fountain solution blending systems measure conductivity or acidic levels  The Summit
solution does not contain metals, and the acid is buffered so that it will not react to
minor changes.

      Summit developed trouble shooting guidelines for press operators, as shown
below in Table 2.
                                     208

-------
TABLE 2.    Trouble Shooting Guidelinws for Summit Solution 61 SOX
PROBLEM
Scumming - non-image
area of plate accepting
ink
Tinting/Toning - fine ink
particles adhere to non-
image area of plate and
appear on sheet as a tint
Ink Roller Stripping - ink
rollers do not accept ink
Poor Ink Drying - ink
does not dry
Plate Blinding - plate
does not accept ink in all
areas, or has low print
contrast
PROBABLE CAUSE
Running excess ink
Glazed offset blanket
Plate or press condition
Excessive printing
pressure
Ink emulsification
Improper ink and/or
dampening roller setting
Running excessive water
Glazed or worn roller
Fountain solution too
concentrated, pH too low
Fountain solution too acid
Running excessive ink
Water too high
Fountain solution too acid
Running excessive water
Cleaners have dried on
plate
CORRECTIVE ACTION
Reduce ink level
Clean all gum, spray, etc.
from the blanket
Wet hone plate in
problem area; if honed
area continues to take
ink, scumming is caused
by an inking problem on
press; if area stays clean,
scumming is caused by
plate sensitivity
Consult appropriate
press manual; pack to
print
Replace with fresh ink
and reduce water
dampening
Check setting on both ink
and dampener rollers;
adjust if necessary
Adjust ink/water balance,
pH = 4.4 or above
Clean or replace roller
Decrease concentration
Check pH, reduced
fountain solution causing
set-off
Reduce ink levels in
delivery
Adjust water level
Check pH, may have to
adjust concentration
Adjust ink/water balance
Wet-wash plate to
remove cleaners
                                    209

-------
Improper Ink/Water
Balance - difficulty in
maintaining print quality
due to build-up, piling, or
emulsification
Plate Wear - plate breaks
down or sharpens
prematurely
Running excess water
Running excess ink
Running excess water
Fountain solution pH
Reduce dampener
settings
Reduce ink roller settings;
if ink levels are low, may
have to increase solution
by half-ounce increments
Reduce ink roller setting
Check pH and adjust if
necessary
TABULATION OF DATA

      Some information from the press trials were subjective, as the press operators
opinions on product quality indicated success or failure of the test.

      The Summit product worked well with all equipment tested except for an AB
Dick press with a silver plate or paperback plate.

      Table 3 shows a comparison of vapor pressure (the amount of evaporation
over time) for traditional fountain solution  and the Summit solution. Vapor pressure
along with related VOC content,  has been reduced drastically.  Traditional solutions'
are made with butyl cellosolve and other alcohols that are volatile.  A typical
concentrated solution contains approximately 5 pounds of VOCs per gallon.  The
Summit solution tested  below detection levels for all VOCs (using guidelines from Test
Method for Evaluating Solid Waste,  Physical/Chemical Methods, Reference 5  Citation
8240, USEPA, Office of Solid Waste and Emergency Response, Washington  DC
20460, SW-846 Third  Edition, November 1986).                          '
TABLE 3.    Vapor Pressire* Comparison of Traditional Fountain Solution and Summit
            Fountain Solution, and Water
   Traditional Fountain
         Solution
Summit Fountain
    Solution
     56- 108 mm Hg
  < 2.5 mm Hg
16 mm Hg
 Determined using a mercury manometer.  Solutions were equilibrated to 20°C for 15
minutes before making the readings.
      Press test results for the four previously described trials are shown in Table 4.
                                     210

-------
TABLE 4.    Press Test Results
PARAMETER
Density
Dot Gain
(acceptable is
<22%)
Dot Structure
Ink Usage
Set-up Usage
Cleaning
Dry Time
TRIAL 1
Brighter, more
gloss
Less than 1 8%
Clean, clear,
circular dots
Used 50% less
black ink and
20% less red
ink than the
last time the
job was run
Used 50
sheets to
adjust press &
density; past
set-ups were
approx. 35-40
sheets
Press required
more initial
cleanup, but
once clean,
was able to
maintain
Sheet dried
faster and
could be cut &
folded sooner
TRIAL 2
Good clear,
high gloss
21-22%
Good, clean,
sharp edges
Used 20% less
black ink; 10%
less red,
yellow, and
green inks
Achieved
proper density
within 300-350
feet
Fountain
solution
started clean,
but the ink
would feed
back onto the
roller,
scumming on
the boxboard
after 10,000
feet. Although
the roller was
cleaned, the
scum would
return.
Faster
TRIAL 3
High gloss,
sharp colors,
good printing
19%
Clean, clear
dot, sharp
edges
Reduced 8-
10%
Ran 45 sheets
before density
& cleaning
were correct
Increasing
concentration
of red ink
corrected
initial problems
Best seen by
pressman
TRIAL 4
Good, sharp
colors, high
gloss
18%
Clean, sharp
edges; good
circular
structure
Reduced 15-
20%
depending on
color
60 sheets
(rollers were
not cleaned
well)
Once rollers
were cleaned
several times,
cleanliness
was
maintained
Improved
                                    211

-------
Customer
Comment






















Results were
good, but the
printer wanted
to ensure long
term results.
Printing plates
started to
oxidize after 4
months. The
plant is not
humidity
controlled, and
the printer
,« i
reported
problems
during the
summer
months with
other
products, as
— M
well.




The product
had many
good features,
but needed to
clean better so
as not to
redeposit on
the roller.
















Press operator
liked the
product, and
although it
was sensitive
to changes, he
felt he could
work with it.
















Customer
liked the
product
because it
eliminated the
20% IPA he
was using.
After using
the product
on his own for
several days,
the printer
had problems
with
scumming &
toning. The
printer
returned to
using alcohol,
and reports
he will try the
Summit
product again
when time
permits.
      Summit Resource Management conducted fifteen trials with similar success.
Customers who continued to use the Summit product had problems if they assumed
the product would work "on  its own" and did not work within the sensitivity parameters
of the product.

      Test result variances did not seem to follow any predictable trends.  Local
conditions, such as water hardness, press subtleties as fine-tuned by an operator  and
pressman skill and attention  to detail influenced the  quality of the printed product.'

      In general, the printers liked the Summit product.  Quality was better, gloss was
greater, dot gain was less, drying time was less, and ink use was reduced.  The same
or better quality product could be printed without  alcohol or toxic ingredients found in
traditional fountain  solutions.

Cost/Benefit Analysis

      The initial cost to purchase one gallon of the Summit fountain solution is
approximately twice as much as for traditional products.  However,  using the Summit
product realizes a savings when the other costs of using IPA are considered  as
shown in Table  5.
                                      212

-------
TABLE 5.   Cost Comparison of Summit Fountain Solution and Traditional Product

Summit Fountain
Solution
Traditional Fountain
Solution and I PA
Costs
Fountain Solution
Black Ink
Alcohol
$25/gallon
$3/pound
$0
$12/gallon
$3/pound
$2.50/gallon
Requirements per Total Gallon of Mixed Fountain Solution
Ounces per Gallon
of Water
Ounces of Alcohol
Ounces of Ink
4
0
1.4
6
19(15%)
2
Savings Review
Material Cost
Alcohol
Ink
Total Use Cost
$0.78
$0
$0.28
$1.04
$0.56
$0.375
$0.375
$1.31
      These savings do not include hazardous waste disposal or water treatment
costs associated with the use of traditional solutions.


CONCLUSIONS

POLLUTION PREVENTION ASSESSMENT

Incentives

      The Summit fountain solution reduces pollution by eliminating IPA and heavy
metals found in traditional fountain solutions. When correctly used, the Summit
product enables the printer to obtain equal or better results in the quality of the
printed product.

      Summit fountain solution also reduces the need for wastewater treatment plants
especially if used with less hazardous inks.  For example, the Summit solution has
been tested successfully with soy inks.
                                    213

-------
Barriers

       Press operators must modify their procedures to successfully integrate the
Summit solution into their process.

       New "dry plate" technology is currently being developed that would eliminate
the use of fountain solutions. Although printing quality is excellent using this new
process, costs are high, and new presses are required.
                                       214

                                    >U.S.  GOVERNMENT PRINTING OFFICE: 1 994-5 50-00 1/OO 1 95

-------