United States
                  Environmental Protection
Risk Reduction
Engineering Laboratory
Cincinnati, OH ,45268
                  Research and Development
EPA/600/SR-93/086   July 1993
? EPA       Project Summary
                   Ink  and  Cleaner  Waste
                   Reduction  Evaluation for
                   Flexographic  Printers
                  Gary D. Miller, William J. Tancig, Michael J. Plewa, and Paul M. Randall
                    The report summarized here de-
                  scribes the technical and economic ef-
                  fects that occurred when a flexographic
                  label print shop changed the type of
                  ink and cleaning agent used in the
                  shop. The changes were undertaken as
                  the best way to eliminate all hazardous
                  materials. The company's management
                  mandated the switch out  of  concern
                  for its employees and with the inten-
                  tion of limiting possible future waste
                  liability. Hence, the traditional  alcohol-
                  based inks and alcohol solvent clean-
                  ing agents gave way  to water-based
                  inks and an aqueous cleaner.
                    From a technical point of view, there
                  is general agreement in this shop that
                  the water-based inks yield better qual-
                  ity labels. Labor is reduced largely be-
                  cause the water-based inks are more
                  easily removed from the pans, rollers,
                  and plates. Ink splashes and spills are
                  also quickly removed by sponging ei-
                  ther with water or with the aqueous
                    As a result of these process modifi-
                  cations, solvent emissions to the plant
                  air have been reduced about 80%. The
                  toxicity  of the gaseous  and liquid
                  wastes has also been reduced approxi-
                  mately 90%. Hazardous liquid wastes
                  have been eliminated while wastewater
                  sent to  the  sanitary  sewer  has in-
                  creased. Solid wastes  have remained
                  relatively unchanged.
                    The water-based inks brought about
                  major savings because most of the liq-
                  uid wastes do not now require disposal
                  as hazardous wastes. The ink wastes
                  are presently acceptable at the local
public waste treatment plant The clean-
ing towels and wipers are now either
rinsed within the plant or sent to a
commercial laundry; formerly, they had
to be labeled as hazardous and segre-
gated for special disposal. Though un-
treated  ink washes are acceptable to
the waste treatment plant, the company
has chosen to filter theirs through a
special  absorbent to remove all color.
The used absorbent is acceptable in
the local landfill.
  For this company, these changes in-
volved  no  capital expenditures other
than the absorbent unit. With the levels
of various alcohols evaporated during
printing now greatly reduced,  the em-
ployees enjoy  a cleaner and healthier
plant environment.
  This  summary was developed by
EPA's  Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings  of the WRITE program
demonstration that is fully documented
in a separate report (see ordering in-
formation at back).

  A Waste Reduction Innovative Technol-
ogy Evaluation (WRITE) (with the Illinois
Department of  Energy and  Natural Re-
sources  [ENR] and the U.S. Environmen-
tal Protection Agency [EPA]) was designed
to (1) quantitatively compare the volume
and toxicity of any  waste generated dur-
ing printing and released as gaseous, liq-
uid or solid waste before and after switch-
ing to water-based inks and a detergent
cleaner and (2) determine the economic
                                                                   Printed on Recycled Paper

effect of modifying a traditional printing
  This project evaluated flexographic print-
ing of labels in commercial quantities. The
participating  firm  was  a  narrow-web
flexographic printer, MPI Label Systems,
Inc., of University Park, IL MPI Label Sys-
tems' plant is in a modern, one-story, clear
span  building. Its printing capabilities are
based on several narrow-web flexographic
presses each capable of printing one to
six colored inks.
  Several  years ago, management di-
rected the company's eight plants to elimi-
nate the use of all hazardous and toxic
materials. The incentives included a de-
sire to eliminate employee exposure to
any liquid or gaseous hazards resulting
from handling these materials and an equal
desire to minimize the possibility of future
litigation resulting from use of such  sub-
  The decision forced each plant to sub-
stitute water-based inks for alcohol-based
inks,  and at the same time, change from
alcohol solvent cleaning  agents to aque-
ous type cleaning agents.
  Water-based inks were substituted for
those formulated with alcohols since the
water-based inks were already available
on the commercial market. Implementing
this change, however, required extensive
cooperation between the  ink manufacturer
and the printing plant because of different
printing conditions, changing customer re-
quirements, and the paper stock require-
ments associated with the water-based
   Because water-based inks are easier to
remove or  clean with  aqueous  agents
when wet,  a terpene-type (d-limonene)
cleaner was initially tried. Later, a  dilute
aqueous solution of detergent proved to
be even easier to handle, odor free, and
less expensive. As cleaning agents are
Introduced to the commercial market, MPI
Label Systems routinely tests them to de-
termine if they can improve their process
and be even safer for the environment.  ,

Printing Background
   Most printing operations function in  a
similar manner. Ink of the desired color
and viscosity is supplied to the  printing
press where a rotating  roller continually
removes a thin layer of ink from the reser-
voir.  This thickness is carefully and pre-
cisely controlled. Ink is then applied to the
printing plate, followed by transfer to the
printing stock. In the case of flexography,
the plate is flexible and is wrapped around
a separate roller. Customarily, when print-
ing labels, a continuous sheet of  paper
receives an  ink imprint  of the image on
the plate. When more than one  color is
involved, a separate plate is used for each
color; each plate must be in registration
for all the colors to be properly placed and
  Paper stock for most labels is a two-ply
material. The top ply, on which the labels
are printed, normally has its bottom sur-
face coated  with an adhesive  that ad-
heres to the  silicone-treated top surface
of the bottom ply. Immediately after the
plate applies  ink to the label stock paper,
the web moves  through a heated drying
area (70°C for this  plant's press).  Each
dry label is  then pressured  by  a  steel
cutting die that physically severs the top
ply of label stock but does not remove the
label  from the surrounding  blank stock.
Moments after a label is severed from the
surrounding stock by the die cutter, the
print-free waste paper stock is peeled away
from the bottom ply and collected  on a
separate roll destined for incineration or a
landfill. The  labels,  still adhering to the
bottom ply,  go  to another roll  and are
processed for shipment to customers.

Waste Reduction
  Wastes are generated at most stages
of the printing process. Ink wastes  result
when  the reservoir, the various rollers,
and  the printing plate are cleaned at the
end  of a run. Excess ink in the  reservoir
can  be collected for reuse, but the other
ink quantities removed during  cleaning
generally remain as waste. MPI Label Sys-
tems keeps a 50-gal barrel of water adja-
cent to each  press.  During cleanup  when
water-based  inks are  used, soiled  parts
and cleaning towels are rinsed in this wa-
ter to remove ink residues. At the end of
the week, barrels of rinse water are trans-
ferred to an  ink splitting device that ab-
sorbs the various ink pigments on a cellu-
lose-based porous  material; the nearly
clear filtrate  passes through for disposal
in the sanitary sewer, with permission from
the local water treatment plant.  The pig-
ment-colored cellulose is accepted at the
local landfill along with  paper wastes from
the print line. Some paper wastes  accu-
mulated during  setup operations are un-
avoidable, although an experienced printer
is usually able to minimize them. Excep-
tional amounts of waste labels are, how-
ever, occasionally generated during the
production of multicolor labels because of
color registration difficulties. The adhesive-
based, print-free stock surrounding each
label is collected on a separate waste roll.
   Most printing processes begin with a
photographic negative.  Developing the
negative generates  a number of  chemical
wastes that usually  require special  treat-
ment for either recycling or disposal. The
photographic chemicals used in produc-
ing the master negative account for small
amounts of waste, though the silver is
usually recovered. At least for flexography,
the plate is developed after exposure to a
bright light and by washing away the un-
exposed areas with a solution.
  In  nearly every step of the printing pro-
cess, some volatile chemicals are released
into the air. In some cases, this occurs as
a result of process design; in  other  in-
stances, it occurs simply because of the
volatile  nature  of the materials. These
volatiles can  range from water to various
alcohols,  plastic ; thickeners, homogeniz-
ers, and chemical diluents. Each ink  loses
volatile  material as a result of storage in
the open.ink reservoir; during rotation on
the various rollers that determine the film
thickness to  be applied to the plate; as
lost material on the plate; and during the
heated  drying phase. The ink  ultimately
reaches a rub-off-free state after the ma-
terial on which it is printed passes through
the heated drying zone. In addition to vola-
tile losses associated with the inks, adhe-
sive  and solvent molecules evaporate from
the adhesive-coated label surfaces  as a
result of simple exposure to the air.  As
with  many chemicals, only a  few  mol-
ecules per cubic meter can, in some cases,
create breathing .distress for hypersensi-
tive individuals, and the sensitive individu-
als must leave the contaminated  area.
Cleaning agents • used on the press will
also evaporate into the air. These clean-
ing solvents can range in type from  water
with  small concentrations of detergents to
various organic chemicals.

  The project compared the volume and
toxicity  of any gaseous and  liquid wastes
produced by the printing  process before
and  after switching to water-based inks
and  an aqueous cleaner and then deter-
mined the economics of such process
   Laboratory measurements estimated the
solvent loss  from  the inks by measuring
the portion of the inks that evaporate and
the  materials used in the plant and by
basing  the calculations on the composi-
tion  of the inks. Laboratory measurement
of solvent loss by evaporation for each ink
was used to estimate the percent volatiles.
The  amount of ink on waste labels Was
estimated by subtracting the total number
of labels printed from the number of  labels
acceptable as product. These results were
compared with the weight of volatiles in
each ink as reported on i:he material safety
data sheets. The  alcohol-based ink con-
tains six volatile components; four of them

are alcohols, and ethyl alcohol and iso-
propyl alcohol are present in the largest
amounts. By comparison, the water-based
ink contains four volatile components, with
most of  the volatiles being water (65%)
and  isopropyl alcohol (5%). Some of the
water, about 24%, is bound to the resins
and  does not evaporate  on drying. Both
the solvent cleaner previously used and
the new detergent cleaner contain over
97% volatiles; again the detergent cleaner
is mostly water. The amount of ink and
other materials disposed of as liquid waste
was determined gravimetrically,  using  a
special rotating apparatus that exposed a
continual and new thin film of ink to  a
constant, low-pressure air jet while the
entire apparatus was kept at 70°C, for two
printing runs.  This approach  minimized
burning,  charring, or spattering of ink as
its viscosity continued to increase during
solvent evaporation. This particular test
was run in triplicate; samples achieved
constant weights, ± 5 mg.


Emissions from Inks and
  Because MPI Label Systems no longer
uses alcohol-based inks nor permits them
in the plant, the actual emissions of these
inks  during in-plant  runs was not mea-
sured. The  percent solids in each ink (and
indirectly, thereby, the amount evaporated)
was determined gravimetrically in the labo-
ratory. These measurements (Table 1) in-
dicate the proportion of the alcohol-based
ink that evaporated was  48%  compared
with 56% to 62% for the water-based inks.
  To .estimate emissions that would have
resulted from using solvent-based inks,  it
was  assumed that the same amount of
solids (or ink retained on the labels after
drying) would have been used for printing
the labels with alcohol-based inks as was
used in  printing  with  water-based  inks.
Then the total amount of ink that would
have been used  and the weight evapo-
rated were calculated  by  using  the per-
cent loss factor determined in the labora-
tory. Since a smaller percentage of  alco-
hol-based ink was lost to  evaporation,
more solids per  gram of ink would  be
applied to the labels than with the water-
based ink. Thus,  less total alcohol-based
ink was  assumed to  be used  and less
total weight of components would be lost
via evaporation. The in-plant measurement
results are shown in Table 2. The water-
based inks-use data were obtained during
two evaluated runs: One run  was approxi-
mately 55,000 labels;  the  other about
250,000. In each case, the total weight of
ink materials added and  the weight of
materials remaining at the end of the run
were measured. The  difference was the
weight of material that was assumed to
be either evaporated to the shop air, dried
on the labels, or wasted.  The  portions
evaporated and retained as solids were
calculated with the use of the laboratory
data presented in  Table 1.
  The press operators at MPI Label  Sys-
tems estimated that about the same total
amount of either ink is required for a job.
Thus, this emission analysis is conserva-
tive for the alcohol-based ink. For com-
parison,  the volume of alcohols evapo-
rated during a printing run was calculated
from the known formulations based on the
total  amount of ink used during each print-
ing run. The laboratory evaporation loss
results also agree quite well with the total
volatile component data shown on  the
material safety data sheets.
  Concentrations of each volatilized com-
ponent in the  plant air were calculated
Table 1.  Weight Loss Data from Laboratory Evaporation at 7CPC
Material Initial wt.,
Water-based 12.17
black ink
Water-based 12.05
green ink
Water-based 12.22
purple ink
Alcohol-based 15.38
black ink
Detergent cleaner 12.56
Average of triplicate runs
Dry wt., Wt. loss,
gm gm
5.30 6.87
5.04 7.01
4.66 7.57
7.99 7.38
0.28 2.22
Std. dev.,
 based on the volume of air in the shop
 area and the air exchange  rate (Table 3)^
 These concentrations are  the  levels ex-
 pected during continual operation of the
 press  assuming all six print stations are
 being  used to apply ink.  Ammonia and
 dimethanolamine levels were highest for
 the water-based inks. For the alcohol-
 based inks, isopropyl  alcohol, methyl al-
 cohol, n-propyl alcohol, ethyl alcohol, ethyl
 acetate, and Varnish Makers and Painters
 (VM&P)  naphtha levels were higher.   :

 Liquid Wastes
   The amount of liquid wastes generated
 from printing  with the water-based inks
 was measured during the two printing runs.
 These wastes (consisting of ink left in the
 reservoir, and on rollers,  gaskets, and
 plates) were produced during cleanup at
 the end of each run. For the green labels,
 24.6 g of ink and 44.3 g of cleaner were
 used. A more experienced operator only
 produced 56.4 g of liquid waste from prirtt-
 ing the purple labels.
   The concentration  of ink and  cleaner
 components estimated to  be present in
 the wastewater  from  the facility  (before
 treatment)  is shown  in Table  4. These
 concentrations were  estimated  assuming
 that each day the ink pans were all in Use
 and cleaned out into a 50-gal tank, every
 week the tank was discharged,  and most
 of  the waste resulted from use of the
 cleaners. From the water-based  inks, over
 85% of the liquid waste was water. Other
 constituents of concern were acrylic resin,
 azo pigments, and isopropyl alcohol. Con-
 centrations of constituents from the alco-
 hol-based inks, if they had been disposed
 of in the  same manner, were estimated to
 be much higher. Toluene, isopropyl alco-
 hol, acetone, resins,  pigments,  and ethyl
 alcohol were the major constituents of con-
   Before implementing the  use of water-
 based ink and detergent cleaner, the sol-
 vent-based waste ink was disposed of as
 a  hazardous waste and was thus mani-
 fested. No liquid ink wastes were  sent to
 the sanitary sewer before the use of wa-
 ter-based inks. Although the total amount
 of liquid  solvent-based waste manifested
 in  a year was on company records, this
 information was considered proprietary and
 was not made available for the  purposes
 of this project. It was not possible to di-
 rectly measure the amount of alcohol-
 based ink and cleaner wastes that would
 have been  generated from printing  runs
similar to those evaluated with the water-
based  inks. Company officials  reported
that in their experience,  the amount  of
solid and liquid wastes generated are es-

Table 2.  In-Plant Ink Use and Estimated Emissions
                          Green Labels
                                       Parole Labels
Water-based   Alcohol-based
    ink            ink
Total Ink used,
Ink solids on
evaporated, g
Table 3.  Estimated Concentration of Air Emissions, \ig/L
                          Purple Labels
                                         Green Labels

Dimethyl 1.1
Water 58
Methyl alcohol
Ethyl alcohol
Ethyl acetate
VM&P naphtha
* Not applicable.
sentially the same for both ink types. The
main difference,  however, is that  liquid
wastes from the water-based inks did not
have to be  disposed of as hazardous
wastes for this project.

Evaluation of Changes in
  Toxicity reduction evaluations  for the
Ink and cleaner wastes were accomplished
with the Degree-of-Hazard scheme devel-
oped and used by the Illinois Hazardous
Waste Research and Information  Center.
The equivalent  toxic concentration was
calculated for four printing scenarios: (1)
green alcohol-based ink with solvent-based
cleaner; (2) purple alcohol-based ink with
solvent-based cleaner;  (3) green  water-
based ink with detergent cleaner;  and (4)
                         purple  water-based  ink  with  detergent
                         cleaner. For estimated emissions to  the
                         air,  the alcohol-based inks and cleaners
                         had relative toxicities about 10 times higher
                         than those for the water-based emissions.
                         The same reduction was found in the liq-
                         uid waste comparison. In actuality, no liq-
                         uid wastes were discharged to the sewer
                         when  the company  was using alcohol-
                         based inks and cleaners. Therefore,  the
                         relative toxicity of these wastes increased
                         from zero. Offsetting  this is the fact that
                         the  undiluted alcohol-based wastes were
                         previously sent to a landfill in the waste-
                         containing drums.

                         Economic Analysis
                           The approximate annual savings at MPI
                         Label Systems resulting from the ink and
                         cleaner change is estimated  by the plant
manager to total at least $16,500 (Table
5). There is essentially no difference in
raw material costs'for the inks and clean-
ers. The overall productivity of the plant
has increased, but the economic value pf
this increase could  not  be determined.
Annual waste disposal and handling ac-
count for at least a  savings of $15,000.
The facility saves about $500 each year
because of a lowered insurance premium
based  on improved working conditions.
Savings because of new wiping materials
(nondisposable) equals about $1,000 an-
nually.           '•

Conclusions  ;
  The results from the change to water-
based  ink and cleaner at MPI Label Sys-
tems are both definite and significant:
  • Solvent emissions to the plant air have
    been reduced by at least 80%.
  • Toxicity of these emissions has gone
    from potentially hazardous to humans
    to essentially harmless.
  • Solid waste generated and destined
    for local landfills has been noticeably
    reduced in volume and is no longer
    classified as hazardous. This  waste
    is mostly  bu|k  paper for which  a
    recycling system is being sought.
  Additionally, these changes neither in-
curred capital costs nor increased opera-
tional expenses. Rather, the plant annu-
ally enjoys a significant saving associated
with reduced waste disposal, insurance,
and  cleaning  material costs.  The  shop
manager at MPI Label  Systems contends
that associated benefits include
  1. Water-based jnks are easier to clean
    from pans,  plates, and rollers when
  2. Water-based ink waste is more easily
  3. Water-based ink spills are easier to
    clean up when wet.
  4. Waste going to a landfill is no longer
    classified as hazardous, reducing MPI
    Label Systems' long-term liability.
  5. Expensive  solvents are not  required
    for cleaning.
  6. Employees  are enjoying a  cleaner,
    safer work environment.
  7. Customers are receiving a  better
  8. Corporate  concerns  regarding
    hazardous materials are minimized.
  The full  report Was submitted  in fulfill-
ment of CR-815829  by Illinois Hazardous
Waste Research and Information Center
under the sponsorship of the U.S.  Envi-
ronmental Protection Agency.

  Table 4.   Estimated Concentrations of Components in Wastewater,

                            Purple Labels
i Components
i Dimethyl
: Pigment
Methy alcohol •
Ethyl alcohol
Ethyl acetate
VM&P naphtha
' Water*
it c-g^c/a 	
 * Not applicable.
 * Constituents of cleaners
, Table 5.  Economic Summary of Savings Using Water-Based Inks and Cleaners

 Parameter	          Savings

 Water-based inks
; Printing speed
' Raw Materials
. Waste disposal and handling
Aqueous cleaners
 Raw materials

Overall Savings
. Insurance liability
; Inventory
 Wiping materials

Annual total
Approximately 10% faster
Minimum annual savings = $10,000
Minimum annual savings = $5,000
Approximately $500/yr
Annually at least $1,000

At least $16,500
                                                                              •ArU.S. GOVERNMENT PRINTING OFFICE: 1993 - 750-071/80016



Gary D. Miller and William J. Tancig are with Hazardous Waste Research and
  Information Center, Champaign, IL 61820. MichaelJ. Plewa is with University
  of Illinois at Urbana-Champaign Institute for Environmental Studies,
  Champaign, IL 61820.
Paul M. Randall is the EPA Project Officer (see below).
The complete report, entitled "Ink and Cleaner Waste Reduction Evaluation for
    Flexographic Printers," (Order No. PB93-191286; 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
       i  EPA
   PERMIT No. G-35