United States
                     Environmental Protection
 Risk Reduction
 Engineering Laboratory
 Cincinnati, OH 45268
                     Research and Development
 EPA/600/SR-94/144   September 1994
<&EPA       Project Summary
                     Waste  Reduction  Evaluation  of
                     Soy-Based  Ink  at  a  Sheet-Fed
                     Offset Printer
                     Beth Simpson, Pamela Tazik, Gary Miller, and Paul Randall
                       This Waste Reduction and Innova-
                     tive Technology Evaluation project fo-
                     cused on the use of soy-based inks as
                     a substitute for petroleum-based inks
                     in sheet-fed offset printing. The goal of
                     the study was to evaluate the waste
                     reduction and economic effects of us-
                     ing soy-based inks in place of the pe-
                     troleum-based  inks traditionally used
                     in sheet-fed offset  printing. The main
                     potential  environmental  benefits
                     claimed for soy-based inks are that they
                     are a renewable resource and emit less
                     amounts of volatile hydrocarbons dur-
                     ing the printing process. No published
                     studies verify the claim of reduced air
                     emissions  or quantify other environ-
                     mental aspects of using soy-based inks
                     such  as the amount of liquid wastes
                     generated and cleaners required. Other
                     reported benefits of soy-based inks are
                     that they are biodegradable in landfills
                     and are more conducive to cleanup with
                     degradable and less toxic cleaners than
                     are petroleum-based inks.
                       This Project Summary was developed
                     by EPA's Risk Reduction Engineering
                     Laboratory, Cincinnati, OH, to announce
                     key findings of the program demon-
                     stration that is fully documented in a
                     separate report (see ordering informa-
                     tion at back).

                      The impetus for developing soy-based
                     inks initially  came from the oil shortages
                     of the 1970's that threatened the supply
                     of petroleum-based chemicals. Now, in-
                     creased  emphasis on improving worker
                     safety and  reducing environmental emis-
sions is motivating the printing industry to
seek cleaner technologies. In the 1980s,
soy-based inks were first used in offset
web  presses for printing newspapers,
mainly for presumed environmental ben-
efits. In the late 1980s, soy-ink formula-
tions were marketed for sheet-fed offset
  Sheet-fed soy inks are defined as those
that have a minimum of 20% soy oil by
volume. The soy oil replaces petroleum
oils in the ink vehicle and varnish compo-
nents. Use of the soy  oil affects color,
drying, and other operating characteristics
of the inks. Soy-based inks used in sheet-
fed presses still contain  at least 10% pe-
troleum oils. Research  is continuing to
increase the proportion  of soy and other
vegetable-based oils in these ink formula-
tions while maintaining satisfactory print-
ing characteristics.
  The Office of Printing Services (OPS)
at the University of Illinois is the operating
partner in this study. They are considered
a medium size, in-plant  printer and gross
about $4.5 million annually. They  agreed
to enter into this cooperative project be-
cause of their commitment to active  re-
search participation and  out of a desire to
be a model environmentally  aware print
shop. They were one of the first sheet-fed
offset presses in the United States to use
soy-based inks.

  Data for this study was collected during
a full-scale  print run on a Miller  TP104
Plus' six-color press in July 1992 at the
OPS. The print job evaluated  was a 4400
sheet, six-color, work-and-turn. Only four
                                                                      Printed on Recycled Paper

of the six colors were included in the study
because the type of ink used for two col-
ors was not changed between the runs.
All inks included in the study were manu-
factured by the same company, Handschy,
Inc. of  Bellwood, IL*  In-plant measure-
ments consisted of weighing the contain-
ers of inks, blanket cleaners,  and roller
cleaners used during makeready and print-
ing at each press unit before and after
each print run. Cleanup rags were weighed
before and after each print run. Wastes in
the wash-up trays were weighed at each
press unit at the end of each run. Samples
of each type and color of ink and each
cleaner used were analyzed at the same
temperature as the press for total solids
and volatile content  at  the Hazardous
Waste Research and Information Center's
(HWRIC) Hazardous Materials Laboratory.
  The operators at OPS use the same
cleaners for  both  petroleum-based and
soy-based inks. They had previously tried
aqueous and less  volatile organic clean-
ers and had not found  any satisfactory
substitutes. Problems identified with some
of the cleaners they tested were that they
required more labor, dried  much slower,
and sometimes left a film on the presses
that had to be removed  in an additional


Ink Use
   During the two print runs studied,
makeready for the petroleum-based ink
run was considerably more involved than
that for the soy-based  run.  This was
caused by excess moisture in the original
batch of paper and not by differences in
the inks. This difference  in  the two runs
affected ink and blanket cleaner  usage
and waste paper generated. The excess
moisture in the paper caused it to curl and
wrinkle during feeding into the press. This
problem was not a result  of the type of ink
being used for that run but because the
paper had been stored  in high  humidity
conditions before the first run. Because of
this, about 25% more pages were printed
using petroleum-based inks, which makes
a comparison of inks and cleaners used
and waste  generated between  the two
runs difficult. These differences were com-
pensated for  to the  extent possible in
evaluating results  of  in-plant  measure-
   The  study results showed that the pe-
troleum-based  ink run  required  larger
quantities of three of the four colors of ink
than did the soy-based ink print run (Table
1). The amount shown in the g/100 sheets
printed column compensate for the differ-
ence in the number of sheets printed dur-
ing makeready. The % difference column
compares the amount of petroleum-based
ink used with the amount of soy-based ink
used. For these  specific print jobs, in to-
tal, approximately  17%  more petroleum-
based  inks were  used  on a per sheet
printed basis.

Cleaner Use
  The amount of blanket cleaners  used is
shown in Table 2. Almost 46% more blan-
ket cleaners were  used during the petro-
leum ink run  (3,455.4g  compared  with
2,368.0g). Some of this difference  can be
attributed to the  larger amount of blanket
cleaners used during the longer makeready
that occurred during this run. The amount
of blanket cleaner used during makeready
was not recorded separately from that used
during  press  cleaning at the end of the
print run. The amount of additional clean-
ers used for that purpose during the print-
ing run could not be directly measured; it
 Table 1. Quantity Inks Used
         would not necessarily be directly propor-
         tional to the amount of paper  printed.
         Therefore, an equivalent comparison can
         not be made between the amount of blan-
         ket cleaners used for each print run.
           At the end of each printing run, roller
         cleaners were  used to clean  up the
         presses. Their  use was  not affected by
         differences in makeready, but there was a
         difference in  the way the presses were
         cleaned at the end of the two runs. At the
         end of the petroleum ink run, the ink in
         two of the stations was not changed. Also,
         since the same colors were used in the
         next run, each ink station was not cleaned
         as thoroughly as it was at the  end of the
         second  run with soy inks. The amounts of
         each roller cleaner used for  each color
         and type of  ink are shown in Table  3.
         Overall, about 30% more roller cleaners
         were used for the soy inks. This was ex-
         pected  since the stations were  cleaned
         more thoroughly at the  end of the  soy-
         based ink run.  Both inks appeared to re-
         quire approximately the  same  amount of
         cleaners and effort to remove from the
         presses. On  average in typical practice,
Ink Color
Ink Type
Ink Used
Ink Used
 Table 2. Quantity (g) of Blanket Cleaners Used
 Ink Type
                               Cleaner Name
                            Cleaner Used (g)
 ' Mention of irada names or commercial products does
  nol constitute endorsement or recommendation for

Clean Quick

Clean Quick


 Table 3. Quantity of Roller Cleaners Used at Each Color Station
Ink Color
Ink Type
the amount of cleaner used for the two
inks would be expected to be about the
  At the end of the petroleum-based ink
run, 663.8 g of ink and  cleaner mixed
waste was in the washup trays. By com-
parison, there was 604.5 g at the end of
the soy-based ink run. This approximately
10% difference can be  attributed to the
extended makeready at  the beginning  of
the petroleum-based ink run. On average
in typical practice, the amount of cleaner
usage for the two inks would be  expected
to be the same.

Volatile Components
  All inks had less than  6% volatile com-
ponents, and there were significantly less
volatile components in the soy-based inks
than in the petroleum-based inks. The pe-
troleum-based inks had an average  of
about 4.6% volatile components compared
with an average of about  0.8% for the
soy-based  inks.  Thus, on  average, the
soy-based inks only had  about 17% of the
amount of volatiles as the petroleum-based
inks. The cleaners  contained more than
97% volatile components except for one
roller cleaner that contained about 88%.
  Based on these  laboratory results and
on the  amount of inks and  cleaners used
(Tables 1 and 2), the amounts  of solids
and volatiles in the inks and cleaners were
calculated  (Table  4). The amounts  of
volatiles were used to estimate the total
mass of  air emissions. Over 90% of the
liquid wastes originated from the inks. As
for the volatiles, over 99% originated from
the cleaners in both cases.
  Because of the difficulty of directly mea-
suring the amount of emissions from the
many sources on the press, the printed
papers, and the waste containers, a worst
case assumption was made that air emis-
sions were equal to the volatile content of
the materials used. Thus,  for the petro-
leum-based ink run, 3,930.15 g were cal-
culated to be evaporated compared with
3,004.30 g for the soy-based ink run. This
was about 31%  more emissions. As de-
scribed  above,  even  after adjusting for
differences in makeready, about 17% more
petroleum-based ink was  used. In addi-
tion, the petroleum-based inks had a higher
volatile  content (4.6% compared  with
0.77%). Just in terms of  ink used, there
were 0.39 g/sheets of  volatile  emissions
from the petroleum inks compared to 0.23
g/100 sheets for the soy  inks. This is
about 70% more emissions to  the air for
an equivalent amount of printing with pe-
troleum inks.
  As  shown in  Table 4,  the  mass  of
volatiles estimated to be emitted from the
inks is less  than 1.0% of the total  mass
estimated to be emitted from the inks and
cleaners. Since most of the air emissions
were from the cleaners, less than 1% over-
all reduction in air emissions resulted from
using the soy-based inks. The  longer the
print run the greater the reduction in vola-
tile emissions from using soy inks..

Liquid Components
  The two main liquid wastes from  the
printing  press  were  from washup  trays
and inks and cleaners on used rags. None
        of these materials were spilled during the
        two runs. To clean the various rollers and
        blankets, several rags were saturated with
        each cleaner and were used to wipe down
        the  press. As a result, the rags contained
        some highly volatile inks and cleaners. An
        undeterminable portion  of these cleaners
        and inks ended up in the washup  trays.
        Rather than determining the rapidly chang-
        ing  amount of liquids on the rags immedi-
        ately at the end of the print run, the rags
        were dried and the amount of dried  solids
        on  the  rags was  weighed. About 60%
        more  solids were  on the  rags  used  to
        clean the presses after the petroleum ink
        run. Since more inks and cleaners were
        used in that run, apparently much of these
        materials ended up on the cleanup rags.
          The amount of liquids  in the washup
        trays  from this  study was measured di-
        rectly at the end of the  print runs. Almost
        10% more liquids were generated in the
        washup  trays after the petroleum ink run.
        This liquid was a mixture of inks and vari-
        ous cleaners. The amount of each ink and
        each cleaner that ended up in these trays
        and the rags (less the amount evaporated
        and on printed product) could not be di-
        rectly  measured.
          The amount of each cleaner  used  to
        saturate the rags was  not  metered out
        precisely by the press  operators so that
        differences  between the two runs in the
        amounts  used were mostly the result  of
        operator variability when  they poured
        cleaners on  the rags and not the results
        of the type  of ink being cleaned. We ob-
        served no reason for there to be  a consis-
        tent difference  in the  amount  of  liquid
        wastes generated by the two types of inks
        studied.  Differences resulted from opera-
        tor variability, extended makeready during
        the petroleum ink run, and more  thorough
        cleaning conducted at the end of the soy
        ink run.  It can not be concluded that one
        ink generated more liquid wastes than the
Table 4. Solid and Volatile Contents of Inks and Cleaners Used (g)
                              Petroleum Print Run
                                                               Soy Print Run
Ink Solids
Cleaner Solids

 Total Solids

Ink Volatiles
Cleaner Volatiles

 Total Volatiles







Solid Waste
  During makeready for the petroleum-
based ink run,  1375 waste  sheets were
generated. The reverse side of 100 waste
sheets were reused during makeready for
the soy-based  ink print run. This is the
amount  of solid waste generated directly
during these print runs. No other contain-
ers of solid wastes were observed. Addi-
tional solid  waste that may be generated
in printing include trimmings and excess
number of  pages printed to compensate
for losses that  may occur during folding,
binding  and any other  final preparation
steps. Because the soy  inks  generally
spread further  (by an average of  about
17%), less used ink containers may be
generated with those inks. The amount of
the other solid wastes would  be unaf-
fected by the type of ink  used.  In  this
case, the difference in makeready waste
resulted from moisture in the paper at the
beginning of the print run  and not from
difficulty with using the inks.
   Press alignment and ink metering  was
automated for the press used in this study.
Based on observations of manually oper-
ated presses at this and other facilities,
automation or efficient press set-up, rather
than the choice of ink, results in less solid
waste being generated  during printing.
Once operators are familiar with the use
of either the petroleum or soy inks, then
the amount of solid waste generated  dur-
ing makeready will  not be noticeably dif-
ferent. In both  cases, more waste paper
might be generated on some jobs be-
cause of difficulties  in obtaining accept-
able colors or other print quality factors. In
this regard, neither ink appears to have a
clear advantage over the other.

   The main cost factor considered  was
that of  raw materials. No equipment ex-
penditures were required by OPS to switch
from using petroleum to soy inks. There
was a difference in the amount of labor
involved in the  two print runs but that was
due to  the problems with  paper at the
beginning of the first run and the fact that
cleanup was more rigorous  at the end of
the second run. Neither of these differ-
ences was due to  the type of ink being
used. The operating  conditions of the
press, such  as temperature and speed,
were the same for each ink so there was
no difference in overall rate of production
or utilities used because of the type of ink
used. Insurance, monitoring requirements,
reporting and  recordkeeping,  and  permit
requirements are the same for each type
of ink. Any differences in future costs for
remediation or property damage would be
minimal. Employee  health costs would
slightly favor the use of soy-based inks
because of  reduced employee exposure
to breathing released chemicals.  All the
quantified cost factors are for the Miller
TP 104 press only and  not for the entire
facility at OPS.
  The purchase prices  of  raw materials
(inks, cleaners, and paper) for this print
job are approximately the same for both
types of ink. Generally soy inks cost about
10%  more.  The average purchase  price
for both the soy inks and petroleum inks
is about $8.00/lb (or $0.018g). Actual costs,
depending on the color,  range from about
$4.00 to $12.00/lb.  Costs do vary,  how-
ever, when the amount of materials used
for petroleum  inks is compared with the
amount of  materials for soy inks. On  an
equivalent basis of  ink used during  print-
ing and typical makeready, the overall cost
savings in ink used for the soy-based  ink
print run was estimated  to be $1.17. This
cost difference is not a major factor in  ink
selection. Higher costs  are  involved in
printing including labor, materials such as
paper, equipment amortization,  and utili-
ties.  The portion of total  printing costs
allocated to inks is usually very small.
When adjusted to the amount of  accept-
able  product  printed, the difference be-
tween cleaner costs for the petroleum and
soy-based  ink runs for this print job was
only  5  cents.  This  difference would  be
greater if the cleaning after the petroleum-
based ink  run  had  been more thorough.
Because the  same  cleaners were  used
for both inks and on average there would
be no difference between the two inks in
the amount of cleaner needed, costs for
cleaning would not be expected to differ.

   Overall,  about 17% less soy ink com-
pared with petroleum ink was used for the
print job studied. Because less amounts
of volatile hydrocarbons are emitted from
the soy-based inks,  air emissions were
less. Since the inks and cleaners  continu-
ously evaporate during use and the waste
inks and cleaners end up mixed together
in washup trays and on the cleanup rags,
it was difficult to determine the proportion
that evaporated during the actual printing
process. Ultimately, most if not all of  the
volatiles in  the inks and cleaners will ei-
ther evaporate, be  discharged to  the wa-
ter (such as from a commercial  laundry)
or be containerized and sent to a landfill.
Some of these emissions will occur at the
print shop,  some perhaps after shipment
of the printed product, and some at waste
management facilities.  The proportion of
volatilization that will  occur in any one
place will vary depending on how the waste
materials are managed. For comparison
purposes,  it was assumed in  this study
that all the air emissions occurred at the
print shop.
  An important factor that influences the
amount of  waste generated per  unit of
production  is the length of the print run.
Some uses of  ink (and resulting wastes
generated)  in the printing process are fairly
consistent among print jobs. For instance,
the amount of  ink  required to coat the
rollers  and plates for each revolution is
relatively consistent. A certain  amount of
waste is expected for every print job, but
a higher proportion  of waste is produced
per raw material input on shorter print
jobs, such as  the  one monitored here.
With longer print jobs, a greater percent-
age of the ink  used would be  applied to
acceptable product.
  To most printers, there  are  little or no
perceived differences in the cost of using
soy-based  inks because the raw material
cost is slightly higher or equivalent to pe-
troleum-based inks. The cost difference is
more than  offset  by the fact that the soy
inks spread almost 20% further and  do
not release as much volatile organic chemi-
cals. Less  tangible  benefits of using soy
inks (company image  as  being environ-
mentally friendly, improved employee  re-
lations due to a  perceived healthier working
environment, and customer preference for
products) are major factors considered by
printers. If  customer response  to product
quality is negative,  printers will not adopt
a new ink. At OPS, their experience has
been that customers find the quality of the
product to  be acceptable,  and  many pre-
fer to have their job printed with soy-based
  A company also must consider the will-
ingness of  its  workers to switch  to and
use soy inks  and  the  time it takes for
operators to become skilled in using these
inks. Understandably there may be some
excess waste  generated  during  such a
change. At OPS, there was some initial
resistance to the new inks. Formulations
of soy  inks in  sheet-fed  offset presses
were new.  In the past few years, these
formulations have improved and the press
operators at OPS have gained the experi-
ence necessary  to produce high  quality
images. Presently,  the employees prefer
to use  soy-based inks on this particular
press. A major reason for this is that the
soy inks  are very  similar to  petroleum-
based inks to  work with and  clean from
presses. If an alternative cleaner was used
that took the operators longer  or required
more effort for the soy inks  then  there
would be considerable resistance and cost.

  In addition to using soy-based inks, there
are other strategies  printers can  use to
reduce wastes from  the sheet-fed offset
printing. Since most emissions  were from
use of the cleaners, these emissions could
be  reduced if aqueous or  less  volatile
cleaners  were used. The press operators
at OPS have not found any of the alterna-
tive products they tried to be acceptable.
It should  be kept in mind that the use of
less volatile cleaners may increase liquid
  Another strategy to reduce waste would
be to reduce the use of cleaners and inks.
This can  be done in two ways; both  may
involve additional capital for equipment.
One way  is to recover or reuse waste inks
and cleaners.  Installing  a solvent still to
recover chemicals from  the washup tray
wastes and possibly reusing those chemi-
cals in formulating cleaners  might be fea-
sible. Some large printers have reportedly
adopted this approach.
  A second way to  reduce  cleaner and
ink  usage  would be to purchase auto-
matic blanket washers and ink  roller train
systems. Literature from one manufacturer
of this equipment claims that as little  as
220 to 275 ml of cleaning solvent can  be
used  during a print run when the auto-
mated systems are  in use. Although this
would  not  eliminate the need for hand
cleaning, it could  greatly  reduce the
amount of cleaner  used and, thus, the
emissions from the cleaners.  It could also
reduce employee exposure to potentially
harmful substances. An automatic blanket
washer could reduce cleaner  use  by up to
90%.  It should be  noted that on some
presses where these systems have been
installed the operators actually increased
use of the cleaners because, even during
a print run, the blankets can be  cleaned
without stopping the press. Automatic ink
handling  systems allow this type of press
to be operated with about 80% less ink in
the fountain.
  The automatic blanket washer and ink
handling  systems also reduce makeready
and cleanup times, saving the press crew
time and exposure to volatile  compounds.
At OPS, it  was estimated  that  such a
system would have a payback of about 3
yr. Most cost savings would be  expected
in the cost  of  cleaners and in  reduced
cleanup time.
  Overall, soy  inks have some  environ-
mental and  other  advantages for sheet-
fed offset printers. The main  environmental
advantage with soy inks is that  they re-
lease less than 20% of the  mass of vola-
tile  organic chemicals compared  to
petroleum inks. The soy inks also spread
about 15% further,  which offsets the small
difference in cost that currently exists. Re-
portedly,  recycling  of soy ink printed pa-
per also  has some advantages. In this
study all other factors including makeready
time,  appearance of printed product, and
cleanup effort were essentially equivalent
between the two types of inks.
  The full report was  submitted  in fulfill-
ment of Contract No. CR-815829 by the
Hazardous Waste Research and  Informa-
tion Center, University of Illinois at Urbana-
Champaign under the sponsorship of the
U.S. Environmental Protection Agency.



 Beth Simpson. Pamela Tazik and Gary Miller are with the Hazardous Waste
   Research and Information Center, Champaign, IL 61820.
 Paul Randall is the EPA Project Officer (see below).
 The complete report, entitled "Waste Reduction Evaluation of Soy-Based Inks
     at a Sheet-Fed Offset Printer," (Order No. PB95-100046/AS; Cost: $17.50,
     subject to change)  will be available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield, VA 22161
        Telephone: 703-487-4650
 The EPA Project Officer can be contacted at:
        Risk Reduction Engineering Laboratory
        U.S. Environmental Protection Agency
        Cincinnati, OH 45268
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268

Official Business
Penalty for Private Use
   PERMIT No. G-35