77-316

ILLINOIS POLLUTION CONTROL BOARD

April 30, 1987

IN THE MATTER OF:

)

VOLATILE ORGANIC MATERIAL

)

 R82-14

EMISSIONS FROM STATIONARY

)

SOURCES:

 RACT

III

)

PROPOSED RULE

 FIRST NOTICE

OPINION AND ORDER OF THE BOARD (by B. Forcade):

This matter comes before the Board

 as part of a regulatory

proposal

 initially filed by the Illinois Environmental Protection

Agency

 (“Agency”)

 on June 30,

 1982,

for

 the control of organic

material emissions from selected industrial categories

and

generic sources.

 The particular proposal

 that is the subject of

today’s Opinion and Order regulates organic material emissions

from one of these industrial

 categories, heatset web offset

lithographic printing.

 Thirty—one hearings have been held,

to

date, regarding the entire R82—14 regulatory proposal.

 A number

of these hearings have specifically addressed the heatset web

offset lithographic printing category.

 An economic impact study

(EcIS) was prepared specifically addressing this category

 (Ex.

71)

On August 10 and

22,

 1984,

 the Board proposed regulatory

language and a supporting opinion,

 respectively,

 for First Notice

(hereinafter, the first First Notice).

 The first First Notice

contained elements of the original Agency proposal, as well as

language and modifications submitted by the Printing Industry of

Illinois (P11).

 Public comments received during the first First

Notice period cited many problems with the proposed rule and P11

specifically requested an additional hearing

 (P.C.

54,

57

&

62).

 On May 30,

 1985,

 the Board, noting

 the confusion and

controversy associated with this category, acknowledged that the

first First Notice rule needed

 revision and that the existing

record needed

 to be supplemented,.

 The Board proposed

 a second

First Notice

 (hereinafter the second First Notice)

 for the

purposed of generating comments and criticisms and authorized

additional hearings.

On September 10,

 1985,

 the hearing officer posed

 a series of

questions and requested that participants respond at hearing.

Hearings solely addressing

 the heatset web offset category were

held on April

 1 and

2,

 1986,

 in Chicago.

 On September

22,

 1986,

the Department of Energy and Natural Resources

 (DENR) filed

a

letter

 indicating that further economic impact assessment would

not be undertaken by DENR for this particular category of rules,

as

 a heatset web offset EcIS was already

 a part of the Board’s

77-315

—2—

record

 (P.C.

 87).

 Final comments were received through September

29,

 1986.

This

 is one

 of a series of Board actions directed at

establishing emission controls

 to achieve attainment of the

National Ambient Air Quality Standard

 (NAAQS)

 for the pollutant

ozone

 (03).

 Ozone is not emitted from tailpipes or smokestacks

like other pollutants, but

 is formed

 in the atmosphere by the

action of sunlight on nitrogen oxides

 (NOx)

 and hydrocarbons

(HC).

 This mechanism, which leads to ozone formation,

 involves a

series of photochemical reactions.

 NOx and HC are, therefore,

called ozone precursors.

 The amount of ozone formed

 in the

atmosphere

 is a function not only of the concentration of

NOx and

HC,

 but also of the meteorology,

 in particular

 the amount and

intensity of sunlight.

 Ozone is a seasonal pollutant,

 reaching

its highest concentrations on warm, sunny summer afternoons.

The

ozone season

 in Illinois extends

 from April through October.

The strategy for controlling ozone has been

 to reduce

 hydrocarbon emissions, which are the primary ozone precursor,

to

the atmosphere.

 These hydrocarbons are termed “volatile organic

materials”

 (VOM)

or

 “organic materials”

 (OM)

 in Board

regulations.

 This regulatory proceeding

 is one of a series that

implements reasonably available control technology

 (RACT)

 for the

control of hydrocarbons from existing major stationary sources

emitting greater than 100 tons per year.

 The implementation of

RACT

 in non—attainment areas

 for ozone

 is required

 as a part of a

federally approvable state implementation plan (SIP)

 under

the

federal Clean Air Act

 (CAA)

 (42 U.S.C.

 7401 et seq.)

 Section

172 of the

CAA

requires that RACT be implemented at existing

stationary sources

 in the non—attainment areas of those states

needing an extension from the 1982 deadline until

 198.7

 to achieve

the air quality standard for ozone.

 Illinois

 is such

 a state,

having requested the extension

 in its 1979 and 1982 SIP.

The definition of RACT

 is contained

 in 40 CFR 51, along with

the requirements for

 a federally approvable SIP.

 However, the

specific parameters

 of what constitutes reasonably available

controls,

 and, therefore,

 the levels of control which the states

must adopt

 to insure that RACT is implemented, are not contained

in federal regulations.

 Instead,

 the United States Environmental

Protection Agency (USEP~)publishes

 a series of documents

entitled “Control Technique Guidelines”

 (CTGs).

 Each of the

CTG5, which are summaries of industry specific case studies,

contains the means

 and the degree of control which

 the USEPA

requires the state

 to adopt categorically as part of its SIP in

order

 to have an acceptable

 SIP.

 Failure

 to adopt rules

identical to those presented

 in the CTGs,

 or other ones

demonstrated by the individual state as comparable, can mean that

the state will have an inadequate SIP, which

 in turn,

 can trigger

the sanction provisions of the CAA found at Sections 110,

 113 and

176

 (42 U.S.C.A. 7410,

 7413,

 7506).

 While

 the mandate for

77-316

—3--

sanctions

contained

the

CAA,

the

mandate

adopt

the

CTGs

or otherwise demonstrate a state rule

 to be comparable

 is not.

It

 is not even contained

 in the federal regulations, but instead

articulated

the

“General

Preamble

for

Proposed

Rulemaking

and Approval of State Implementation Plan Revisions for Non—

attainment Areas”

 (44 FR 20372).

This federal policy statement

 includes yet another

requirement which is relevant to this rulemaking.

 The USEPA

allows the states until

 the January after one year from the

finalization of a CTG to adopt either the “rules”

 contained

therein,

 or comparable rules,

 if sources covered by that

particular CTG are within a state’s non—attainment areas.

 Also

of interest

 is the unstated policy of the USEPA

 to publish draft

CTG5.

 Draft CTGs are informally distributed

for

 the purpose of

generating comments.

 These comments are often incorporated

in

final CTG publications.

 Presumably,

 state adoption of rules

comparable to draft CTGs

 is not mandatory.

 A draft CTG has been

issued

 for the heatset web offset industrial category, but was

withdrawn or terminated

 by letter, dated March 22,

 1982, from

USEPA Deputy Administrator John Hernandez (Exs,

 29(e),

 24(o)).

The significance of this will be discussed further

 in Section

1,

below.

The proposed regulation of the heatset web offset industrial

 category has been one of the most complex and controversial

regulatory proceedings

 in recent memory.

 This

 is due

 to the

multiplicity of technical and legal issues that have arisen

in

the course of this, now, five—year proceeding.

 Consequently,

it

is necessary to separately address each issue

 in what is, hope-

fully,

 a logical progression.

 The general categories are as

follows:

1)

 necessity and rationale for regulation of the

heatset web offset category;

 2) description of heatset web

offset printing process and potential emission sources;

3)

 scope

of regulation

—

 fountain solutions and ink solvents;

 4) geo-

graphical applicability of the proposed regulations;

 and 5)

content of regulations

—

 technical and economic

 issues associated

with control options.

 Necessity and Rationale for Regulation of the Heatset Web

Offset

Industrial

category

specific

rules

that

 are not CTG—based but

are, nonetheless,

 RACT.

The criterion for determining whether the heatset web offset

industry

needs

RACT

regulated

whether

not

sources

emitting

over

100

tons/year

exist

 areas designated non—

attainment for ozone.

 Emissions less than 100 tons/year would be

below the strict legal

 threshold established

 in the

CAA.

 Whether

or not such emissions are de minimus

 for the purposes of air

quality planning for attainment

 is a separate

 issue.

There are two separate potential sources of emissions from

the heatset web offset printing process:

 VOMs in the fountain

solution and organic material emission from heated ink

solvents.

 While there

 is disagreement between the Agency and P11

as to whether

 ink solvent emissions should be regulated at all,

there

 is no dispute that VOMS in the fountain solution are

 legitimate subjects of regulation

 if emitted

 in sufficient

quantities.

 Information prepared and submitted

 by the P11

in

post—hearing comments shows both isopropyl alcohol

 (isopropanol)

usage and emissions (the primary VOM in fountain solutions)

and

ink solvent usage and emissions for heatset web offset printing

facilities

 in non—attainment areas

 (P.C.

 82, Table A—C).

 Table A

of this survey shows two facilities

 in non—attainment areas with

isopropanol emissions greater

 than 100 tons/year.

 Thus, even

if

the Board proposed regulatory scope only included fountain

solution VOM emissions, major stationary sources exist

 in non—

attainment areas.

 These figures do not take into account the use

of isopropanol substitutes which are also VOM.

 Consequently,

some form of RACT regulation is an absolute requirement under the

CAA.

 The regulatory choices that remain are generic controls now

proposed

 in R86—lB or rules specific to the heatset web offset

industry.

 At this stage

 in the proceeding,

 the Board believes

that

 it is best to propose category specific rules

 in this R82—l4

docket for imposition of RACT, rather than subject this category

to generic controls.

As

 a general matter, category specific

rules that account for unique aspects of an industrial process

 are preferable to generic regulations.

 Comments are specifically

requested on this issue.

A second, correlative

 issue,

 is whether

 the levels of

control prescribed in the terminated draft CTG constitutes RACT

for the heatset web offset industry.

 This

 is a separate issue

from whether category specific rules are legally required as

a

77-318

—5—

consequence of the CTG.

 The Board’s second First Notice and the

Agency’s most current proposal are based on the terminated draft

CTG.

 However,

 as the Board stated

 in the May 30, 1986, Opinion

proposing

 the second First Notice:

“The

 Board

is

not

 advocating

 this

 proposed

language but

 is using this second First Notice

opinion

and

 order

as

 a vehicle

for

 reopening

the

record

this

category

 and outlining the

unresolved

issues..The

new

language

will

provide

starting

point

develop

achievable

and

 reasonable

 rule.”

 (R82—l4,

RACT

III,

Opinion,

May

30,

1985,

pp.

1—2)

The

Board

 finds that the regulations based on the withdrawn draft

CTG

are

not

necessarily

RACT

for

this

category

and

that

the

Board

not

 bound

 to promulgate regulations equivalent

 to those

contemplated

 in this document.

 The Board must promulgate rules

that, based

 on the record, represent PACT and are technically

feasible and economically reasonable pursuant to Section

 27 of

the Act.

 The issue of whether ink solvents will be included

in

these PACT controls will be addressed further

 in Section

3,

below.

2.

 Heatset Web Offset Lithography

 Process and Emission Sources

“Heatset” refers

to

 a class of web—offset lithography which

uses a heated dryer

 to solidify or set the printing inks by

driving off excess solvents from

 a printed surface.

 “Offset”, as

used

 in the lithographic printing industry, refers to the blanket

cylinder which transfers ink from the plate

 to the surface

 to be

printed.

 “Web” refers

 to the continuous roll—fed printed

substrate or paper.

Each printing unit of

 a press has a series of vertically

arranged rollers and cylinders above and below the web.

 These

roller/cylinder systems draw either water—based

 fountain solution

or solvent based

 ink from wells.

 Maintaining the distinction

between image and non—image areas to be printed

 is done through

chemical means.

 The non—image areas are receptive to water,

or

fountain solution.

 The image areas are water repellent and oil

or

 solvent receptive,

 so that the ink stays on the image areas.

The

 fountain solution and the inks are transferred by complexly

arranged rollers to the plate cylinder.

 The image

 is then

transferred

 from the image plate to a rubber covered blanket

cylinder and then to the web.

 The infeed section of the press

allows

 the rolls of paper

 to be mounted, aligned, unwound and fed

through the press.

In

 a typical process—color heatset web offset lithographic

printing press,

 each printing unit simultaneously applies a

single color

 to both sides of the web.

 Together all printing

77.319

—6—

units can overlay colors for

 a full color

 image without drying

between printing

 units.

 After

 the printing web leaves the last

printing unit,

 it enters the dryer.

 The most common type of

 dryer

 is a high velocity, hot air blower.

 Air temperatures can

be as high

 as 500°F. Much

 of the heated

air

 is recirculated,

with only enough being discharged to prevent the buildup of

explosive solvent vapors.

 The web leaves the dryer with surface

temperatures between 266°Fand 329°Fand travels over an assembly

of driven steel drums with chilled water circulating

 through them

which cool the web to a maximum 860?.

 This cooling,

in

combination with the evaporation of the ink

 in the dryers,

prevents the ink from transferring to adjacent sheets when the

printed web is cut, folded and stacked

 (R. 667—668, 2713;

Ex.

29(e)).

There are two types of materials, fountain solutions and ink

solvents, used in heatset web offset printing

 that result

in

organic emissions from the process.

 The fountain solutions used

are typically composed of an etchant, such as phosphoric acid,

gum arabic, a dampening solution, such as isopropanol, and

water.

 The etchant is often purchased

 in a premixed concentrate

that contains the etchant, gum arabic, mineral

 salts

and

 a very

small quantity of solvent.

 These solvents are VOM

 (R. 4044).

Isopropanol, which

is

 a VOM,

is

 a commonly used dampening

agent.

 High print quality

 is often attributable to the level

of

isopropanol used.

 Generally,

 a higher level

 of isopropanol

in

the fountain solution results

 in better print quality.

 Typical

isopropanol

 usage ranges from 15—25 percent of the fountain

solution.

 Automatic dampening systems usually maintain

a

20

percent level, while manual make—up systems range

 from 15—25

percent.

 While alcohol substitutes are available, these

substitutes are all VOM.

 However,

 the alcohol substitutes are

generally less volatile than isopropanol

(R.

 4046; PC.

 62).

The

feasibility of replacing isopropanol with lower volatility

substitutes

 is limited and

 a minimum five percent isopropanol

is

necessary for dampening systems using older, less flexible

rollers

 (R. 666—671,

 4001; P.C.

 62).

Ink solvents,

 or ink oil, are hydrocarbons comprised of

mixtures of narrow cut petroleum fractions having an

 average

molecular weight of about 206.

 C1~and C22 hydrocarbons have

been identified

 in ink solvents ana

 a commonly used solvent has

C12 and C16 hydrocarbons.

 The composition of the hydrocarbons

could include saturated alkanes, unsaturated olefins and

aromatics.

 The solvents boil within limited temperature

ranges.

 Frequently,

 the boiling ranges identify the ink

 solvent.

 For example, Magie 470 oil has

 a boiling range of 462

F

to 516°F. Most ink formulations contain 35 to 43 percent, by

weight, hydrocarbons

(P.

 4030—4032,

 4040).

 Two major types of

ink solvents are used

 in heatset inks.

 One series of solvents is

a severely hydrotreated variety of the other.

 Magie

Sol

47

is

the hydrotreated version of the Magie

 470 oil.

 Hydrotreatment

77-320

—7—

results

converting

the

unsaturated

olefins

and

aromatics

into

saturated compounds.

The

 ink solvents used

in

 the heatset web offset industry do

not fall within the current regulatory definition of VOM, as the

solvents have vapor pressures less than 0.0019 psia at 70°F.

35

Ill. Adm. Code 211.121

 and 215.104.

 Neither do they fall within

the regulatory definition of “photochemically reactive material”

at 35

 Ill. Adm. Code 211.122.

 The heatset web offset industry

switched

 to these

 ink solvent formulations

 in order

 to be

exempted from the applicability of the existing generic organic

emission limitation of

 8 lbs/hour at 35 Ill. Adm. Code 215.301

(R.

 3990).

Emissions from the heatset web offset printing process

emanate from the printing unit (i.e., the fountains and the

roller/cylinder

 system)

 and the dryer.

 The terminated draft CTG

estimates that 50 percent of the fountain solution emissions

occur

 in the pressroom from the press unit and 50 percent occur

in the dryer.

 However,

 the Agency believes that emissions from

the press unit occur

 in the range of 0.8

to

 25 percent, while 75

to 99.2 percent of the emissions evolve off of the web

 in the

dryer

 (Ex.

 28(g)).

 The emission factor

 for fountain solution

VOM5 is 100 percent, i.e., virtually all VOMs

 in the fountain

solution volatilize and are emitted

 to the atmosphere from both

the printing unit (i.e. pressroom emissions)

or

 the dryer vent.

No

 ink solvents are emitted from the printing unit because

of their low volatility at standard temperature and pressure.

The vast majority of the ink solvent organic emissions that occur

evolve

 in the dryer, which volatilizes the ink solvents through

high heat.

 These emissions are emitted

 to the atmosphere via a

stack from the dryer.

 The Agency contends that all of the ink

solvent emissions that occur,

 occur

 in the dryer

(P.

 3957).

However,

 a very small amount of emissions may come off the web as

 it exits the dryer and travels on the cooling rollers.

 Some

secondary outgassing may occur from an extremely hot web

(P.

3959).

 Some portion of the

 ink solvents

 is retained on the

printed web, or product, and

 is never

 released to the

atmosphere.

Emission factors for the

 ink solvents are difficult to

quantify.

 The terminated draft CTG estimates that 20 percent of

the

 ink solvents remain

 in the web, or product, which would

result

 in an emission factor of

 80 percent.

 P11 contends that

emission factors vary depending on the type of product being

printed.

 Product variables that affect emissions include:

the

relative absorbency of different types of paper, the ratio of

printed

 to unprinted surface,

 the number of colors used and the

thickness of the printed

 ink layer

 (P. 4042).

 These variables

can result

 in emission factors ranging

 from 50 percent to 80

percent

(R.

 4041,

 4043).

 Results of a long—term study conducted

77-321

—8—

by World Color Press,

 Inc.,

 involving 37 printing jobs using a

wide variety of press configurations and web paper,

 found that

the web typically retains 19.96 percent of the ink solvent

applied which corresponds to an emission factor of approximately

 80 percent

 (P.C.

84,

 p. 13).

P11

 in its emission survey used an

average emission factor figure of 70 percent (P.C.

 82).

 Because

of the variability of products produced,

 there is variability in

the amount of emissions,

 which are dependent on the absorbency of

the paper and the amount of ink applied.

 Because of the

variability in emissions it is very difficult to quantify the

emissions with precision.

 The nature of the printing business

is

such that printers cannot control the type of product produced,

as

it

 is done on

 a job—shop basis

(P.

 4043,

 4047).

Many heatset web offset dryer vents are controlled

 in some

manner, either by afterburners or condensers.

 These controls are

necessary,

 in some circumstances, because of opacity and odor

regulations.

 Plumes of condensed

 ink solvent vapors can cause

opacity violations,

 absent controls.

 Odor controls are often

necessary in urban

 areas.

 Consequently, all of the presses

located

 in urbanized non—attainment areas have some form of

control

device

(P.C.

82).

3.

 Scope of Regulation

The main focus of controversy and disagreement

 in this

proceeding has been whether or not the organic emissions

 from ink

solvents should be regulated.

 P11 contends that:

1)

 these

emissions are de minimus

2)

 the ink solvents are not VOM5 as

defined

 in current Board

 regulations;

3)

 a large portion of the

dryer vent emissions quickly condense and are therefore not

available for gas—phase photochemical reactions

 in the

atmosphere; and 4)

 the ink solvent emissions are not

photochemically reactive and should not be regulated.

 The Agency

contends that:

 1) emissions are not de minimus but are

approximately 2000 tons/year

 in non—attainment areas and over

5500 tons/year

 in attainment areas;

2)

 ink solvents are emitted

to the atmosphere by heat volatilization in the dryer;

3)

the

results of the various studies are inconclusive regarding

reactivity;

and

 4) unless specifically excluded from regulation

by USEPA,

 ink solvents should be regulated.

 In support of these

arguments,

 P11 and the Agency have presented testimony and

exhibits regarding volatility, condensation and reactivity of ink

solvents and ink solvent emissions.

P11 presented the results of

 a study conducted by Battelle

Columbus Laboratories (“Battelle Study”) concerning the

volatility and reactivity of commonly used

 ink solvents

in

environmental chamber irradiation experiments

 (Exs.

 22, 39,

101(b); P.C.

 54).

 This project was contracted

 for by the Graphic

Arts Technical Foundation,

 a printing industry research

organization.

 The first part evaluated the volatility of heatset

77-322

—9—

printing ink solvents and the feasibility of conducting tests

within smog chambers to determine their photochemical reactivity

(Ex.

 22).

 The second part evaluated ink solvents reactivity in

comparison with the hydrocarbon ethane

 (Ex. 39).

 A third part

compared recondensed

 ink solvents with “fresh”

 ink solvents

in

order

 to determine

 if the printing and drying process alters

their composition

 in such a way as

 to increase or decrease

reactivity

 (P.C.

 54; Ex. 101(b)).

 Additionally, the third part

extended the work performed

 in the previous two parts and

included experiments on the reactivity of isopropanol, Magie 500

oil and toluene

 (P.C.

 54; Ex. 101(b)).

Task

 1 of the Battelle Report investigated the volatility of

heatset printing solvents

 in order

 to determine the portion that

would be available for participation in the gas—phase reactions

important

 in the photochemical production of ozone.

Two

solvents, MagieSol

 47 and Magie 470 oil, were used

 in the

study.

 Various methods of volatilization were used, one method

being

 found most appropriate.

 Task

 1 demonstrated that it was

technically feasible to proceed and evaluate

 the relative

reactivity of different materials under ratios of hydrocarbons to

nitrogen oxides known to lead

 to ozone formation

 (R.. 755—758).

Task

 1 also found that the solvents were sufficiently volatile

that “virtually all of the oil constituents are available to

participate in gas—phase photochemical reactions”

 (Ex. 22).

However,

 results from Task

 1 do not rule out the possibility that

condensation can occur

 under certain conditions.

 Condensation

is

experienced

 in the field

and

 is evidenced by visual smoke

 (Ex.

111(a)).

 Condensation is primarily a function of concentration

of oils

 in the stack and particulates in the atmosphere that

provide

 a locus

 for condensation.

 Stack gas temperature and

atmospheric conditions also influence condensation.

Unfortunately, the question of exactly how much of the solvent is

available for gas—phase reactions remains unanswered.

 The Task

1

experiments do not cover

 this aspect adequately

 to support

quantification of how much ink solvent

 is available

in

 a gaseous

state and how much condenses.

Task

 1 also focused on possible photochemical aerosol

formation during chamber irradiations.

 The formation of a

photochemical aerosol would indicate that the test materials are

reactive and contribute to the formation of ozone.

The

environmental chamber background air contained

 a high ratio of

hydrocarbons

 to nitrogen oxides (NOx).

 After approximately two

hours of irradiation,

 a photochemical aerosol appeared during the

experiments with Magie 470 oil,

 but did not with those conducted

with MagieSol

47.

 The authors concluded

 that this was due

 to the

aromatic content of the 470 oil, which was assumed

 to be

10

percent.

 Based

 on this assumption,

 they calculated that

20

percent of the oil

 is converted

 to aerosol duringthe

 •two hour

irradiation.

 However,

in

 a subsequent analysis of MagieSol

47

and Magie 470 oil using gas chromatograph/mass spectrometer

77-323

—10—

(GC/MS),

ultra

violet

(UV)

absorption

and

NMR

techniques,

 it was

found that the 47 oil had no detectable level

 of aromatic and

that 470 oil contained,

 at most, one percent aromatic

 (Ex.

110).

 In light of this new understanding of the aromatic content

in these oils,

 it must now be assumed that all

 of the aromatics

and some additional component of the 470 oil

 is photochemically

reactive.

 Using the one percent aromatic content assumption and

carrying out

 a calculation similar

 to the one performed in the

Battelle Study,

 100 percent of the aromatic and

 a portion of the

aliphatic component of the 470 oil would be converted to aerosol

through

photochemical

reactions.

There

are

variety

 of parameters that can be used

to

evaluate photochemical reactivity.

 The Battelle Study identified

eight and chose one, maximum ozone concentration,

 to be used

as

the yardstick for the Task

2.

 One series of experiments was

conducted

 to compare the reactivities of the two ink oils to that

of ethane.

 In some experiments, concentrations were expressed on

a mass basis, that is parts per million as carbon, while

in

others molar concentrations were employed, that is parts per

million by volume.

 In both cases, the oils produced

 a higher

ozone concentration than ethane within the first twelve hours of

irradiation, although ethane eventually generated more ozone when

compared by mass.

 It must be noted that the ratio of hydro-

carbons to nitrogen oxide was 5:1, much higher than normally

 found

 in an urban mixture.

 HC/NOx ratios of 1.5 to 2.0 are

typical

 in urban atmospheres.

In another

 series of experiments,

 ink solvents or ethane was

added

to

 a typical atmospheric hydrocarbon mixture composed of

seventeen hydrocarbons.

 Recalling that part of the purpose of

the second part was to compare the oils’

 reactivity to ethane’s,

in approximately half of this series of experiments, the oils

were substituted

 in place of the ethane used

 in the other half.

When ethane was replaced by MagieSol

47,

 the maximum ozone

concentration dropped

 5 percent.

 When it was replaced with Magie

470 oil,

 it dropped about 13 percent.

So

 this series

demonstrated that replacing ethane with either of the ink oils

results in

 a reduction in the maximum concentration of ozone

formed

 in the first twelve hours

 (Ex.

 39).

In response to comments by Dr. Basil Dimitriades of USEPA,

Research Triangle Park, additional studies were performed to

gather data under conditions that were more realistic

 in terms of

hydrocarbons to NOx

 ratios that exist

 in the atmosphere.

 This

report, which contains

 the results of Task A and

 B, also extended

the work of Task

 1 and

2.

 This further investigation of

reactivity was also performed by Battelle.

 This report used the

data from the Task

 2 Battelle Study

 in the analysis.

 Task A also

included experiments on the reactivity of isopropanol, Magie 500

oil and

 toluene.

 Task

 B involved performing three experiments

to

determine whether printing oils are modified by the printing

77-324

—11—

process

 in a manner that would affect their photochemical

reactivity.

Task A experiments assumed

 that synergistic and inhibitory

effects in multicomponent mixtures can best be represented by

utilizing a matrix of atmospheric organic compounds and that such

a procedure

 is a realistic method for comparing the reactivity of

a test compound such as the heatset oils with a reference

compound (ethane).

 Smog chamber experiments were carried out at

non-methane organic compounds/nitrogen oxides (NMOC/NOx)

 ratios

of 1.5,

 2.8 and

5.

 The authors concluded that the three repre-

sentative

 ink oils, namely MagieSol 47, 470 and 500

 “....are

generally no more reactive than an unreactive reference compound

(ethane).

 One exception

 is the 470 oil at NMOC/NOx of 1.5, where

reactivity of the oil exceeds that of ethane”

 (Ex.

 101(b)).

Task

 B experiments

 investigated whether

 the heatset web

offset printing process alters

 the ink oil

 in such a way that the

oil’s reactivity would be affected.

 The experiments show that

the reactivity of oil emitted from an actual press run was equal

to the reactivity of the same oil which had not been exposed

to

the printing process

 (Ex.

 101(b)).

The USEPA contracted with William P.L. Carter to conduct

a

computer modeling study of the photochemical reactivity of

heatset printing oils

 (Carter Report).

 This study was carried

out at the Statewide Air Pollution Research Center

 (SAPRC)

 of the

University of California

 in Riverside

 (Ex. 101(d)).

 The purpose

of the Carter Report was

 to use a mathematical modeling approach

to study the mechanistic aspects of heatset ink oil reactivity

in

light of the data obtained from the Battelle experiments.

The study consisted of two major

 tasks.

 The first was to

simulate the results of the Battelle chamber experiments based on

current understanding of the chemical reaction mechanisms of the

higher alkanes and thus determine the most appropriate way to

represent the oils

 in model

 simulations.

 The second

 task is

strongly dependent on the outcome of the first task.

 In the

second

 task, box—type airshed model calculations were carried out

to assess the relative contributions to 03 formation from the

addition of heatset oils.

In carrying out the first

 task,

 a number of major

assumptions were

 involved.

 First,

 a choice of 0.6 ppb for the

chamber dependent proportionality factor was made.

 The authors

indicate that this was a best fit.

 However,

 a look at Table

2

shows that the model calculated values for O.~maximum are very

much lower

 (about 40)

 than the O~maximum o5tained

experimentally (Runs 2—16 and 2—7).

 Several other chamber—

dependent parameters are assumed by the authors to be appropriate

for simulating

 the Battelle experiments.

 Second,

 a detailed

mechanism for the NOx—air reactions

 of ethane,

 propane, n—butane,

77-325

—12—

n—pentane,

 iso—octane,

 toluene, m—xylene and

 their oxygenated

reaction products was assumed to represent the reactions of

ethane and

 the components of the urban surrogate used

 in the

Battelle experiments.

 The authors have included comments

in

Table

 1 on why such a representation of the surrogate mix was

used.

 A third and more controversial assumption

 is the

representation of the ink oils.

 The authors used n—pentadecane,

which has a molecular weight close to the average molecular

weight of the ink oils.

 A probable set of reactions for this

 compound are included.

 In addition, m—xylene

 (2—10)

 was added

to the model

 to represent the reactivity of the oils and

to

better fit the data from the single oil component experiments

(Table 2).

 The following discussion relates to this last

assumption.

The aromatics

 in the oils are represented by varying the

amounts of m—xylene added

 to the n—pentadecane.

 It was believed

by the authors that both Magie 470 and Magie

 500 oils contain

approximately 10—12

 aromatics.

 However, as stated earlier and

presented in Exhibit 110,

 these oils may contain no more than 1

aromatics,

 If this is really the case,

 the use of m—xylene to

represent the reactivity is probably not appropriate.

 Then the

addition of m—xylene would simply be an artifact to raise the 03

concentrations predicted by the model.

 The fact that maximum 03

concentrations obtained

 in the chamber experiments using the 47

and 470 oils are not too far apart does suggest that the aromatic

content of the 470 oil

 is not too large.

 From the results of the

model

 simulations of the urban surrogate—NOx experiment and the

urban surrogate with added ethane or printing oils (shown

in

Table

 3, Exhibit 101(d))

 the author’s conclusion that the

representation of the printing oils as n—pentadecane plus

variable m—xylene (2—5

for

 the 47 oil and 5—10

for

 the 470 oil)

is not justified by the data.

 In particular the m—xylene

percentage

(5)

 that demarcates

 the 47 oil

 from the 470 oil

is

not clearly seen in the data.

 Thus the use of this representa-

tion can at best be described as qualitative.

 Quantitatively,

the model requires more refinement.

The second task

 in the Carter Study deals with the assess-

ment of the relative reactivities of ethane and

 the mixtures of

compounds thought

 to represent the printing oils.

 This has been

done by measuring the change

 in O~concentration caused by the

addition of known

 small amounts o~the test compounds (ethane,

mixtures representing the

 printing oils

 or the urban surrogate)

to the assumed existing emissions.

 Two Empirical Kinetic Model

Approach

 (EKMA)

 scenarios and two multi—day with stagnation or

transport scenarios were used for modeling the relative

reactivities.

 Based on results of the modeling, the author

states

 that:

“under

 practically

all

 conditions

 except

the

highest

 HC/NOx

 ratios,

 then n-pentadecane,

 5—

77-326

—13—

10

 m—xylene

 mixtures,

 which,

 based

on

the

chamber simulations,

 is taken

 to represent the

reactivity

of

the

 two printing

 oils most ex-

tensively

 studied

by

 Battelle,

are

 signifi-

cantly more reactive than ethane.”

Further, Carter notes that the

“mixture

 (sic)

 taken to represent the printing

oils are less reactive than the mixture taken

to

 represent emissions

 from

 other

 sources

in

urban

 areas,

 indicating

 that

 these

 oils

are

probably less

 reactive

 relative

 to O~ forma-

tion than most

 pollutants emitted

 info

 urban

areas.”

The authors conclude with

 a discussion of some of the

weaknesses of their assumption of

 a n—pentadecane and m—xylene

mixture

represent

the

ink

oils.

note

the

statement

that

such a representation is “our best estimate of

 a chemical model”

and that it

 is “necessarily highly approximate,

and

 it contains a

number

 of uncertainties.”

 The reaction mechanism for n—penta—

decane

 is based on an extension of models

 for C4—C9 alkanes

because limited data exist

 for reaction mechanisms

 for alkanes

with more than four carbons.

 These points

 in their conclusion

suggest

 a need

 to quantify the uncertainty wherever possible.

The results do indicate that the reduction in the aromatic

content of ink oils can reduce the reactivity to that of ethane.

Another

 important result from the modeling study was the

effect of hydrocarbon to nitrogen oxide

 (HC/NOx) ratios on

predicted daily maximum ~

 concentrations.

 The maximum increase

in 0~above that predicted

 in the base case

 is seen to occur

at

the low to moderate HC/NOx ratios

(4

 to 8).

 However

 the absolute

predicted 03 concentrations are lower

 at the low HC/NOx ratios.

HC/NOx ratios of 1.5 to 2.0 are typical

 in urban atmos-

pheres.

 The ratio of concentrations of ink oils to NMOC is also

expected

 to be low in the atmosphere.

 Not having carried out

computer runs at HC/NOx ratios below 6, the authors extrapolate

from the available data to state that the 03 production

 is less

sensitive

 to added organics at low HC/NOx ratios

 (i.e., below

6).

 Thus,

 evidence of any increase

 in ozone production due

to

ink solvents is likely to be obscured.

P11 contends that the quantity of ink solvent emissions are

 de minimus and should

 not be regulated as a significant source of

ozone precursors.

 This contention is not supported by the

record.

 As previously discussed,

 the CAA provides

 a legal

threshold for regulation of 100 tons/year

 for stationary

sources.

 P11’s own survey on isopropanol

 and ink solvent usage

in non—attainment areas shows that ink solvent emissions are

in

77.327

—14—

the area of 2000 tons/year from the industrial category with

approximately nine facilities emitting over

 100 tons/year of

isopropanol and ink solvents

 (P.C. 82).

 Ink solvent emissions in

attainment areas are approximately 5500 tons/year.

 The estimate

of ink solvent emissions is based on a 70 percent emission

factor, which is favored by P11.

 The record indicates that

higher emission factors are appropriate

 in some circumstances

(Ex.

 29(e);

 P.C.

 84).

 Even with this possibly low estimate,

there are major stationary sources in non—attainment areas,

 thus

necessitating regulation purely based on quantity of emissions.

P11 also argues that not all of these emissions are available

in

the gas—phase

 arid

that

those

that

are

available

are

non-

reactive.

 However, assuming for the moment that ink solvent

emissions are appropriately subject to regulation as ozone

precursors, from a pure quantity of emissions standpoint,

 the ink

solvents are not de minimus.

P11’s second major argument

 is that a large portion of the

ink solvent emissions from the dryer condense from the gas—phase

back

 to the liquid phase

 and are, consequently, not available for

photochemical reaction

 in the atmosphere.

 The record shows that

condensation of

 ink solvent emissions does occur

 to some degree

in the industry.

 Condensation can result in visible plumes of

smoke

(Ex.

23,

111(a)).

consequence,

many

heatset

web

offset

presses

are

controlled

either

afterburners

condensers

order

 avoid violations of the Board’s opacity

regulations

(P.

 3989—3990,

 4151; P.C.

 82).

 However, industry

witnesses admit that this condensation plume formation is not an

automatic occurrence and “in many instances, there are presses

in

different plants where

 the concentrations that we are able to

account for are not adequate to form

 a condensate”

(R.

 773).

Condensation

 is dependent on the concentration of oil emissions

 in the stack, ambient temperature and ambient particulates

 in the

atmosphere, which provide a locus

 for condensation.

Additionally, even when condensation does occur,

it

 is unclear

what

portion

the

emissions

remain

the

gaseous

 state.

 Task

the

 Battelle Study demonstrated

 that ink solvents volatilize

when subjected

 to heat and “virtually all

 of the oil constituents

are available to participate in gas—phase photochemical

reactions”

 (Ex.

 22).

 Task

 1 left unanswered

 the question of how

much of the solvent emissions are available

 for ozone

formation.

 One industry witnesses indicated that there are no

numbers

 in existence quantifying

 the condensation phenomenon,

in

part because the quantity constantly changes depending on

production factors and atmospheric conditions

 (R. 4121—4122).

In summation, while

 the record shows that the phenomenon of

condensation of ink solvent emissions does occur

 in some

circumstances, there

 is little factual support for P11’s position

that a significant portion of ink solvent emissions are not

available in

 a gaseous state

 for photochemical reaction.

By

P11’s own evidence condensation does not occur automatically, the

77-328

—15—

quantitative aspect of condensation is totally unknown and its

occurrence is dependent on fluctuating meteorological and

emission conditions.

 Based on this record, the Board cannot

accept P11’s argument that a significant portion of the emissions

are not available for ozone formation.

 The evidence before the

Board indicates that under certain conditions,

 all of the ink

solvent emissions remain

 in a gaseous state and are available for

photochemical reaction in the atmosphere

 (Ex. 22).

P11 argues that ink solvents, as presently constituted, are

not VOM5 as defined

 in Board regulations at 35 Ill. Adm. Code

211.122 and 215.104.

P11

 is absolutely correct

that

heatset

ink

solvents

not

fall

within

the

current

 regulatory definition of

VOM,

which

written

terms

volatility

 specified

standard

temperature

and

pressure.

This

argument

might

persuasive

this

was

 adjudicatory proceeding construing

existing

regulatory

 language.

 See DuPage Publications v. IEPA,

PCB

85—44,

85—70

and

85—130,

____

P.C.B.

____,

May

1986;

P.C.B.

___,

 August 14,

 1986.

 However, the purpose of the instant

proceeding

 is to first determine whether this industrial category

should be regulated,

 and then,

 if regulation

 is necessary, what

level

control

PACT.

The

Board

liberty

this

proceeding

to

fashion

 regulatory language that will address the

issue of whether or not ink solvents should

be

 controlled.

In

response to this issue, the Agency proposed an amendment to the

definition of VOM that would include ink oils.

 Because of

potential impact beyond the scope of the heatset web offset

industrial category, this proposed amendment was separately

docketed as

 a new regulatory proceeding,

 P86—37.

Regardless of the current definition of VOM,

 the real issue

is whether the ink solvents are emitted to the atmosphere

 in the

course of the heatset web offset printing process.

 Regarding

this particular issue, there

 is little factual dispute that the

high temperature dryers, which

 “set” the inks,

 volatilizes a

 large portion of the ink solvents.

 These volatilized solvents

are emitted through dryer stacks

 to the atmosphere.

 While there

is variability

 in the emission factors,

 a reliable range is

70

to

80 percent

 (P.C.

82,

 84).

 As previously discussed, some portion

of these emissions can condense under certain conditions but that

portion cannot be reliably quantified.

 Thus,

 regardless of the

current VOM definition most commonly used

 in PACT regulations,

organic

 emissions are volatilized into the gaseous state and are

emitted

 to the atmosphere in significant quantities.

 The Board

is not limited

 to using

 the existing VOM definition

in

the

context of these rules and can certainly use the term “organic

materials”

 if appropriate.

 The undisputed

 facts show that

volatilized organic material emissions do result from the heatset

printing process.

P11’s

 final argument

 is that the heatset ink solvents are

not photochemically reactive and should not be regulated.

The

77-329

—16—

Agency contends that ink solvents are photochemically reactive

and that there

 is an insufficient factual basis

 for excluding

them from regulations as ozone precursors.

 The Agency and USEPA

view the evidence generated on photochemical reactivity as

inconclusive.

 At the outset of this discussion,

it

 is apparent

from the studies performed to date that the relative

photochemical reactivity of heatset ink solvents is close

 to that

of ethane.

 Ethane

 is exempted from regulations as an ozone

precursor by both USEPA and the Board because

 it is negligibly

photochemically reactive and, therefore,

 not of regulatory

concern.

 Whether

 ink solvents are more or less reactive than

ethane

 is uncertain.

 Under certain environmental conditions,

ink

solvents are less reactive and, under other conditions, they are

more reactive

 (Exs.

39,

 101(b)).

 Another point that is apparent

from a review of the record

 is that both ethane and

 ink solvents

are photochemically reactive, i.e.,

 they generate ozone under

atmospheric conditions

 (Ex.

 22).

 Very nearly all organic

compounds that are

 in the gas—phase react in the atmosphere to

ultimately form ozone.

 P11’s assertion that the ink solvents are

not photochemically reactive

 is clearly an overstatement.

For regulatory purposes, organic compounds have been

categorized both

 in terms of volatility and reactivity.

The

volatility classification is premised on the concept that only

 organic materials that are volatile at standard temperature and

pressure enter the atmosphere as gases and are, therefore,

available

 for photochemical reaction.

 Of course, organic

materials can be volatilized through heat or pressure

 in the

course of an industrial process.

 This aspect has already been

discussed as

 it relates

 to the heatset ink solvents.

 Organic

compounds have been classified

 in terms of the rate at which they

photochemically react.

 Organic materials that react slowly over

time have been classified as low reactive; organic materials that

react more quickly are classified as reactive.

 Very few

materials are totally non—reactive or inert.

 The choice of

ethane as a benchmark for regulation

 is not

 a purely scientific

or technical decision but is,

 in fact,

 a regulatory decision

which

 is based on the best data available along with other

planning and policy considerations.

 Ethane and certain other

selected materials are excluded from regulation because they

react

 so slowly as to have

 a negligible impact on air quality.

The decision whether or not to regulate ink solvents

 is likewise

a regulatory decision which encompasses

 a review of the available

scientific data, the reliability and certainty of that data,

an

analysis of the potential air quality impact of the emissions and

 the regulatory framework for regulation.

 The issue can be

distilled to this:

 Are the data presented sufficiently

conclusive to support a finding that heatset

 ink solvent

emissions have

 a negligible

 impact on air quality due

 to their

extremely low photochemical reactivity?

77-330

—17—

The P11 relies primarily on the results of the Battelle

Study in support of its position that ink solvents are non-

reactive

 (Exs.

22,

39,

 101(b)).

 P11 argues that the Battelle

Study

 is the only credible evidence

 in the record on

 ink solvent

reactivity and that this evidence shows that they are equivalent

to or less reactive than ethane.

 P11 criticizes

 the findings of

the Carter Report based on alleged errors in certain key

assumptions and methods.

 The Agency maintains that ink oils

participate

 in photochemical reactions in the atmosphere and

that, unless specifically excluded from regulation by

 a final

rulemaking action by USEPA,

 they should be controlled.

The

Agency and P11 agree that the USEPA is undecided on the issue of

whether

 ink solvents are significantly photochemically reactive

and whether

 they should be excluded from regulation.

 USEPA views

the current data as “inconclusive.”

 USEPA continues to view ink

solvents as ozone precursors subject to regulation

 in the absence

of conclusive data.

 No formal decision has been made on the

issue of whether or not to exclude them from regulation.

The results of

 the Battelle Studies do provide some of the

best evidence presently available on ink solvent reactivity under

certain conditions.

 However,

 the results and conclusions that

can properly be drawn are limited.

 Task

1

 of the Battelle Study

shows that ink oils can be volatilized with heat and will remain

in a gaseous state.

 Task

 1 also demonstrates the ink solvents’

ability to photochemically react,

 i.e.,

 formation of

a

photochemical aerosol after

 irradiation.

 Battelle Tasks

2,

 A and

B results show that under various simulated environmental

conditions,

 ink solvent reactivity varies

 in relation to

ethane.

 Under most of the simulated conditions, the solvents

appeared less reactive than ethane.

 Magie 470 oil at NMOC/NOx of

1.5 was more reactive than ethane.

In reviewing the Battelle data,

 the Board must consider

the

reliability of the data and the conclusions drawn from that

data.

 Statistically speaking, there were relatively few

replicate samples from which

 a comparison of the reactivities of

the solvents to ethane could

 be made.

 Except at the NMOC/NOx

ratio of 5.0 which had two runs each for isopropanol and the ink

solvents and four runs for ethane, there

 is essentially just one

run for each compound

 tested at the other NMOC/NOx ratios.

Conclusions drawn from such limited data should be viewed with

caution.

 Additionally, certain of the test conditions were not

standard throughout

 the tests comparing ethane with ink

solvents.

 At the NMOC/NOx ratio of 2.8, the ratio of test

compound

(ethane)

NMOC

was

0.11

while

all

other

experiments

were run at equal molar concentrations of test compound and

surrogate urban mixture.

 Because of this,

 a rather high value

(509 ppb)

 occurs

 for the maximum ozone obtained for

 run A—2 with

ethane and

 the urban mix.

 The tests using

 ink solvents (Pun 2—8,

2—9,

A—b)

resulted

lower

ozone

values.

This

comparison

ethane tends

 to make the solvents look as

 if they are less

77431

—18—

reactive.

In

 fact,

 the test conditions were not comparable.

On

the other hand,

 when equal molar concentrations of ethane and

 surrogate urban mixture are used, at the same NMOC/NOx ratio of

2.8,

 the maximum 0.~produced

 is 378 ppb (Run A—13).

 If the ink

solvents had also been

 tested under

 these conditions,

it

is

possible that they might have produced a maximum 03 concentration

in excess of 378 ppb;

 in which

 case, the conclusion would have

been

that

the

ink

solvents

were

reactive

than

ethane.

fact, this latter conclusion

 is plausible based on the

observation that ethane reactivity decreases faster

 than that of

the ink oils for reductions

 in the NMOC/NOx ratio from 5.0 to

1.5.

The results of the relative reactivity are presented

in

figure

the

Summary

Report

Task

and

(Ex.

101(b)).

Conclusions from this data regarding the reactivity of ink

solvents are possible only if the conditions of the experiment

are also stated.

 The conditions are necessary for reasonable

interpretation since the solvents are more reactive than ethane

under some conditions and less reactive under other conditions.

The only conclusion that can be drawn from the summary results

is

that the reactivities of the solvents are not very different from

 that of ethane under

 test conditions.

 It also appears

 from the

data that

 a reduction in the ratio of test compound

 to NMOC

increases the reactivity of the ink solvent test compounds with

respect

 to ethane.

 Since the actual concentrations of the

heatset ink solvents

 in the atmosphere

 is low compared

 to the

urban mix,

 the data suggests that the oils might be more reactive

than ethane and,

 therefore, produce more ozone than ethane would

under

 likely environmental conditions.

In summary, there

 is ambiguity in some of the Battelle Study

results and inherent limitations to drawing broad conclusions

from environmental chamber test results.

 It is not possible to

exactly simulate actual ambient atmospheric conditions in

environmental chamber experiments.

 The Battelbe results

 show

that ink solvent reactivity

 is dependent on the experimental

conditions.

 Additionally,

 it is not practical to simulate,

in

environmental chamber studies,

 the full range of reaction

conditions which occur

 in the atmosphere, and which affect the

relative reactivity of the materials being compared.

The Carter Report was intended

 to help fill

 in these

informational gaps through computer modeling based

 on Battelle

Study data.

 The Carter Report,

 first, explored the chemical

 reaction mechanisms of the higher alkanes

 in order

 to accurately

represent the ink solvents

 in model simulation.

 Second, Carter

conducted box—type air—shed model calculations

 to assess the

relative contributions to ozone formation.

 Carter made

 a number

of conservative assumptions regarding chamber—dependent para-

meters,

 the mechanisms

 for NOx

 to air reactions representing the

Battelle ethane

 to urban surrogate reactions, and the

77-332

—19—

representation of the

 ink solvents.

 This last assumption

 is the

most controversial and,

 in light of subsequent data, perhaps

erroneous (Ex. 110).

 The actual aromatic content of the test

ink

solvents is much lower than presumed by either Carter or

Battelle.

 Thus,

 in the case of the Carter Report,

 the choice of

in—xylene to represent the reactivity is probably not appropriate

and could artificially raise the ozone concentrations predicted

the

model.

Only limited conclusions can be drawn from the Battelle and

Carter reports.

 The experimental data does not conclusively

settle the reactivity issue.

 The assumptions about the reaction

mechanisms are flawed because of the current lack of knowledge.

GC/MS analysis of sample ink solvents indicate extremely low

 levels of aromatics, much less than previously believed (Ex.

110).

 The Battelle Study concluded that the photochemical

reactions that did occur during chamber irradiations were

attributable

 to the assumed 10 percent aromatics.

 The findings

of

Ex.

 110 undercut this conclusion.

 Some component of the ink

solvent, other

 than

 the aromatics, must be reacting at rates

higher than previously attributed

 to the higher alkanes.

The

assumptions of the Carter Report ink solvent surrogate are also

undercut by Ex.

 110.

 Part of the problem is due to the fact that

ink solvents are not pure compounds, but are comprised of various

components.

 The exact composition of these complex solvent

formulations can vary from lot

 to lot (P.C.

 84).

 Additionally,

not much

 is known about

 the photochemical mechanisms of the

higher alkanes, above

 C10, which comprise

 a large component of

the ink solvents.

 Because of these informational uncertainties,

it

 is difficult to draw conclusions with a high level

of

confidence.

It is necessary to review the regulatory strategy for

control of ozone precursors

 in bight of the uncertainty

surrounding the composition and photochemical reactivity of the

ink solvents.

 Early federal and state efforts at ozone control

focused

controlling

higher

reactive

organic

materials

and

allowed exemptions

 for low (slow)

 reacting organic materials.

This approach, initially adopted

 in California’s “Rule 66”, was

adopted

the

Board

and

now

found

Ill.

Adm.

Code

211.122

 (definition of “photochemically reactive material”)

and

215.301.

 USEPA regulations in this area also allowed

 for a

control strategy of:

1)

 reducing organic material emissions

generally; and

2)

 replacing highly reactive material with lesser

reactive material.

 40 CFR Part 51 Appendix

B.

 Under

 this

regulatory scheme, ink solvents are presently exempt from the

8

lbs/hour level

 of control under 215.301.

Subsequent

 to this first effort at ozone control, the

regulatory strategy changed.

 USEPA’s guidance to the states

indicated that the reactivity concept was useful as an interim

measure only and would not be considered

 a reduction in emissions

77-333

—20—

for

purposes

estimating

attainment

the

ambient

air

quality

standard

for

ozone.

USEPA

severely

reduced

the

category

materials

deemed

not

regulatory

concern

due

their

extremely

low

reactivity

from

what

was

 previously excluded under the “Rule

66”

strategy.

35314

(July

1977).

Only

four

materials

were

excluded

from

 regulation,

 one of which

 is ethane.

 This

listing

has

been

expanded

include

eleven

compounds,

date.

Illinois

adopted

this

approach

its

definition

VOM,

which

excludes

the

eleven

federally

excluded

compounds.

USEPA analysis of available data

and

information

showed

that

very

few

VOMS

are

such

 low photochemical reactivity that they

can

ignored

ozone

control

programs.

USEPA

found

that

many

VOM5

that

were

previously

designated

low

reactivity

 materials

are

now

known

 moderately or highly reactive

 in urban

atmospheres.

Second,

even

 compounds that are presently known to

have

low

 reactivity can form appreciable amounts of ozone under

multiday

stagnation

conditions

can

occur

summer.

35314.

The

Board

finds

that

 the scientific data presented to date

inadequate

justify

exclusion

ink

solvents

from

regulation

ozone

 precursors.

 While the data presented does

show

that

ink

 solvent reactivity is close to that of ethane,

it

only

under

certain

 conditions.

 Additionally, the data

is

too limited

 to draw broad conclusions on ink solvent reactivity

throughout

the

spectrum

atmospheric

conditions.

This

limited

data,

combination

with

the

present

lack

 knowledge on the

photochemical

behavior

the

ink

 solvents, cannot support

regulatory exclusion since ink solvents are emitted to the

atmosphere and they are photochemically reactive.

 While the ink

solvents are generally slower reacting, their emission to the

atmosphere contributes

 to the formation of atmospheric ozone and

is of special concern during multiday stagnation scenarios.

Under atmospheric conditions experienced

 in Illinois and

southeast Wisconsin,

 gaseous ink solvent emissions slowly react

to form ozone.

 Under

 the current regulatory strategy adopted by

Illinois, it

 is appropriate and necessary to control ink solvent

emissions.

Where

 the record before the Board demonstrates that a source

category has substantial emissions of hydrocarbons to the

atmosphere and that those particular hydrocarbons are

photochemically reactive and will probably lead

 to the formation

of ozone under

 usual atmospheric

 conditions, the Board

is

justified

 in adopting technically feasible and economically

reasonable regulations to control

 those emissions.

 The Board

finds that during the heatset printing process,

 ink solvents are

volatilized and emitted

 to the atmosphere

 in a gaseous state and

in quantities that are of regulatory concern.

 While condensation

can occur,

 it has not been shown to significantly reduce the

gaseous emissions.

 Data presented to date shows that ink

77-334

—21—

solvents are photochemically reactive.

 Their rate of reactivity

that

 of ethane but varies depending on experimental

conditions.

unclear

how

reactive

ink

solvents

are

under

actual atmospheric conditions as the existing test data is

limited

and

little

known

about

the

 reaction mechanisms of the

higher alkanes, which are principal components of ink solvents.

Test

data

 does indicate that greater reactivity is exhibited

under conditions approaching probable atmospheric concentrations

of

ink

solvents.

Because

the

data

does

not

show

that

ink

solvents are of such low reactivity to warrant exclusion based on

limited impact on air quality, especially during prolonged

irradiation under multiday stagnation conditions,

 the Board will

establish

PACT

controls

 for both fountain solutions and ink

solvents.

4.

 Geographic Applicability

When the first regulations controlling heatset web offset

printing were proposed as part of the PACT III regulatory

package,

 they were intended

 to apply on

 a statewide basis.

 This

was consistent with

 the strategy undertaken in the PACT

I

 (R 79—

2,

 3) and PACT II

 (P 80—5) proceedings.

 Several years ago, when

these proceedings were completed and RACT III was proposed, much

of the state was designated

 as non—attainment.

 When PACT

 I was

initiated,

 25 counties

 in Illinois were non—attainment for

ozone.

 The rationale for statewide applicability was based on

the pervasive statewide ozone problem,

 the atmospheric transport

ozone

and

ozone

 precursors from sources

 in attainment areas

to

non—attainment

areas,

and

the

need

provide

for

growth

the

SIP

 (R. 40—63).

 At present, many areas of the state have

achieved attainment for ozone and the major non—attainment areas,

with one exception, are concentrated

 in the Chicago and East St.

Louis major

 urbanized areas

 (P. 3204—5).

 Macoupin County is not

 located

in

 a major urbanized area but continues to experience

violations of the NAAQS

 for ozone.

Recent regulatory proposals have focused

 on implementing

PACT in the nine counties that comprise the Chicago and East St.

Louis major urbanized regions and Macoupin County.

 Eight of

these counties are currently designated non—attainment for ozone.

Will

 and McHenry counties are currently designated attainment for

ozone but are part of the Chicago urbanized area.

 The SIP must,

in addition

 to imposing PACT on major stationary sources

 in non—

attainment areas,

 provide

 for ultimate attainment of the ozone

NAAQS.

 To that end, sources

 in Will and McHenry still need

 to be

PACT controlled

 in order

 to ensure adequate emission reductions

because of the transport of ozone

 and ozone precursors from these

geographically contiguous counties.

During the course of the various Agency, Board

 and P11

regulatory proposals for the heatset web offset category, no

participant has raised the

 issue of changing the geographic

77-335

—22—

applicability

in

light

the

current

SIP

strategy.

Consequent-

ly, the Board will limit the geographic applicability of PACT

 controls to the ten counties designated either non—attainment for

ozone or

 that are a part of the Chicago urbanized area.

The

Board

 is cognizant that this action will greatly decrease the

economic impact of emission reduction contemplated

 i.n previous

proposals.

 World Color Press

 Inc. was identified

 in the EcIS as

potentially bearing 62

 of the total statewide cost of the

regulation at four of its facilities located

 in attainment areas

(Ex.

71).

These

facilities

will

not

subject

PACT

limitations that require

 the installation of add—on pollution

control equipment.

 However, the Board will require

 some level of

control of fountain solution VOM on

 a statewide basis.

 This

level

control

will

something

less

than

full

PACT

controls

but will nonetheless limit VOM emissions.

The

rationale

for

requiring

some

level

 of statewide control

based

on,

first,

the

need

maintain

the

current

attainment

status

throughout

most

the

state.

Approximately

eight

major

stationary

sources

are

located

areas

that

are

currently

attainment (excluding Will and McHenry counties which are

considered

part

the

Chicago

urbanized

non—attainment

area).

Total

estimated

organic

emissions

from

these

facilities

range

from

2600

tons/year

5200

tons/year

(Ex.

71).

Many

these

facilities

are

extremely

 large sources of organic emissions to

the

atmosphere.

Second,

emissions

from

these

facilities,

although located

 in attainment areas, can contribute to ozone

in

non—attainment

areas

through

atmospheric

transport

ozone

and

ozone precursors.

 One facility, located

 in Randolph County,

is

contiguous to

 the East

St.

Louis

major

urbanized

non—attainment

area.

 Emission reductions on a statewide basis will help reduce

the

ambient

ozone

and

ozone

precursor

concentration

loadings

that

can

impact

non—attainment

areas.

Content

Regulation

 Level

 of Control

The

PACT

control

options

for

heatset

web

offset

printing

that

can

prescribed

regulation

 are summarized as

follows:

 (1) reduction of VOM in the fountain solution through

reformulation;

(2)

 installation and operation of a thermal or

catalytic incinerator

 to control dryer emissions;

and

(3)

installation and operation of

 a condenser/filter system that

selectively removes ink solvents and other low volatility

materials such as isopropanol substitutes,

 but does not

effectively remove

 isopropanol.

 Ink reformulation

 is not

currently a

PACT

option

 (P.C.

 62).

During the course of this proceeding,

 there have been

numerous

regulatory

proposals

for

 the control of the heatset web

offset printing process.

 At least four separate proposals merit

discussion:

(1)

 the Agency’s proposal which was analyzed

 in the

EcIS;

(2)

 the Board’s first First Notice proposal of August 10,

77-336

—23—

1984;

(3)

the

P11’s

proposal

(P.C.

62);

and

 (4) the Board’s

second First Notice proposal of May 30, 1985, based on

the

terminated

draft

CTG.

This

proposal

has

been

adopted,

with

modifications,

the

Agency

its

 current proposal

 (Ex.

 103)

Certain elements of these various proposals are not technically

feasible or economically reasonable.

 Many of these deficiencies

have

been

raised

hearing

public

comments

and

will

discussed further below.

The

Agency’s

proposal,

which

was

analyzed

the

EcIS,

called for statewide regulation of

 facilities emitting 100

tons/year or more of organic material.

 Three control options

were prescribed:

(1)

 installation and operation of an

afterburner

which

oxidizes

percent

the

organic

material;

(2)

 the fountain solution contain no more than five percent of

volatile organic material and

 a condensation recovery system is

installed and operated that removes at least 75 percent of the

organic materials from

 the airstream; or an alternative control

system equivalent to either of the previous control options.

The

major

problem

with

this

rule

that

limitation

five

percent

VOM

the

fountain

does

not

appear

technically

feasible

for

many

heatset

web

offset

presses.

The

Board’s

first

First

Notice

rule

proposed

August

10,

1984,

applied

statewide

facilities

whose

emissions

VOM

exceeded 25 tons/year.

 The rule required one of three options:

(1)

installation

afterburner

system

which

oxidizes

percent

captured

non—methane

VOM;

(2)

reduction

VOM

concentration

the

fountain

solution

than

five

percent and installation of

 a condensation recovery system which

removes at least 75 percent of VOM5 from the airstream or

reformulation of the

 ink to

 a high solid/low solvent; or

(3)

an

alternative control

 system demonstrated

 to have

 an equivalent

emission reduction efficiency equal

 to either of the first two

options.

This proposal presented

 a number of conceptual problems.

First,

the

proposed

rule

attempted

regulate

only

VOM

emissions

yet

prescribed

control

ink

solvents.

The

various

control

strategies were not equivalent.

 Certain options were not

technically

feasible,

such

the

VOM

content

the

fountain

solution

and

the

ink

reformulation

 option.

The

P11

proposal

 provided for statewide regulation at a 40

tons/year VOM threshold

(P.

 4119).

 Alternatively, P11 requested

a

 40 tons/year/press

 threshold (P.C. 82).

 No justification for

 this level has been provided.

 The P11 rule would require use of

afterburner

which

oxidizes

percent

the

VOM

emissions

presented

the

control

equipment;

(2)

VOM

limitation

percent

the

fountain

solution;

(3)

equivalent

alternative control system.

 The main problems with this proposal

were the exclusion of ink solvents from regulation and control

and the forty tons/year/press threshold.

77~337

—24—

The Board’s second First Notice provided statewide

regulation of sources emitting over 100 tons/year

 of VOM.

The

proposal provided four alternative control strategies:

(1)

 total

elimination

VOM5

the

 fountain solution; or

 (2) reduction of

VOM

concentration

the

fountain

solution

percent

and

installation and operation of an

 incinerator; or

(3)

 reduction of

VOM concentrations in the fountain solution to seven percent and

installation and operation of a condenser/filter system; or

(4)

alternative

emission

 control system equivalent to any of the

first

three

options.

This

 proposal had a number of problems

associated

with

it.

First,

total

elimination

VOM

the

fountain

solution

not

technically

feasible,

nor

limitation

seven

percent.

Second,

the

structure

the

regulation favored the incineration control option.

 Third, the

various

levels

fountain

solution

VOM

which

corresponded

and

triggered application of add—on controls were arbitrary.

All

the

regulatory

proposals

date

 have allowed an

unspecified alternative equivalent control strategy.

 Preliminary

comments from USEPA

 indicate that

such

option

probably

not

federally approvable

 (R. 3898—3901;

 Ex. 110).

As discussed in Secton

 1 of this Opinion, while the Board

is

required

adopt

PACT

regulations

controlling

the

heatset

web

offset

category,

the

specified

level

control

that

PACT

has

not been federally defined.

 Thus,

 the Board

 is at liberty to

define

level

control

that

PACT,

based

 the regulatory

record.

The

regulatory

controls

must

also

technically

feasible

 and economically reasonable as a matter of state law.

Reconciling what is RACT and what is technically feasible and

economically

reasonable

 possible, as the concept of PACT

incorporates elements of reasonableness, cost effectiveness and

technical feasibility and availability of control options.

The Board will propose as a third First Notice regulations

that

hopefully

meet

these

federal

and

state

standards,

based

the regulatory record.

 The threshold for regulation will be 100

tons

per

year

organic

material.

This

threshold

consistent

with

the

CAA

definition

major

stationary

source.

This

will

include

organic

materials

that

are

 considered volatile at

standard temperatures and pressures,

 as well as non—volatile

organic

materials,

such

 the ink solvents, that are volatilized

during

the

printing

 process.

 first control alternative, proposed Section 215.408(a)

(1)

will

require

installation

and

 operation of an incinerator

that oxidizes at least 90 percent of the organic material

present

the

airstream

from

the

dryer.

 This approach will control

nearly

all

 the volatilized ink solvent emissions.

 A majority

of the fountain solution VOMs will also be controlled through the

use of an

 incinerator.

 While

the

 terminated draft CTG estimates

that half of the fountain solution VOM emission occur

 in the

77-338

—25—

pressroom,

 the Agency has presented evidence that from 75 to 99.2

percent of the fountain solution VOM emissions occur

 in the dryer

(Ex. 28).

 Thus, even

 if higher levels of VOM are used under

 this

control

option,

large

fraction

the

fountain

solution

VOM

emissions

 will be captured and controlled.

 This option will

provide flexibility

 in the printing process to accommodate high

quality printing jobs while ensuring a high level of control.

Because the process involves a high heat dryer that volatilizes

the vast majority of fountain solution and ink solvent emissions

directly

into

the

dryer

vent,

capture

efficiency

needed

specified.

 The Board envisions

 a situation where the control

device is directly connected

 to receive the dryer vent airstream,.

thus obviating the need for

 a capture device.

 This will also

obviate the practical problems of specifying a capture efficiency

for

this

particular

 application of control technology.

The

second

alternative

control

 option,

 Section 215.408(a)

(2),

will

include

control

VOM

the

fountain

solution

eight

percent

and

the

installation

and

operation

condenser/filter system that captures and removes at least 75

percent

 of the non—isopropanol organic emissions from the dryer

airstream.

 Condensation recovery systems can effectively remove

ink

solvents

and,

possibly,

low

 volatility isopropanol

substitutes,

but

will

not

 effectively control isopropanob.

Consequently,

it

necessary

reduce

VOM5

 the fountain

solution

 in order

 to control their emission

 to the atmosphere.

The record indicates that fountain solution VOM can feasibly be

reduced to eight percent without negatively impacting print

quality.

 Once again, no capture efficiency is needed or

specified for the condensation control system as

 it is envisioned

that dryer vent emission will be directly routed

 to the control

device.

 A removal efficiency of

 75 percent of non—isopropanol

organic emission from the dryer airstream appears reasonable as

nearly all of the organic emissions will be ink solvents and,

therefore,

 recoverable.

As a separate control requirement, proposed Section

215.408(b) will provide an eight percent VOM limitation

for

fountain solution at facilities located outside the ten counties

 designated either as non—attainment or part of the Chicago

urbanized area.

 This limitation

 is technically feasible,

according to P11,

 and will cost industry nothing.

 The level of

control required by Section 215.408(b)

 is less stringent than

PACT and should be easily met by the eight impacted facilities.

No unspecified alternative equivalent control option

is

provided as

 it would probably not be federally approvable.

P11

 has objected

 to most of the regulations proposed

 to date

as being economically unreasonable and technically infeasible.

First,

 the Board believes that the rules proposed

 today are

technically feasible.

 Many concepts and levels of control

77-339

—26—

advocated by P11 have

been

incorporated

the

rule

such

the

eight percent limitation on fountain solution VOM and the use of

afterburners without a specified capture efficiency.

 Second,

regarding economic reasonableness, the Board believes that the

proposed rule provides flexibility

 in the choice of control

options either through incinerators or fountain solution

reformulation

and

condensation

system.

Both

these

options

are

 cost effective and are compatible with existing industry controls

(R,

4124—4127),

Condensation

recovery

systems

are

identified

the

most

cost

effective

control

option

because

the

revenue

derived

from

the

sale

combustion

recovered

solvent

(Ex,

71).

Reduction

expensive

isopropanol

and

other

fountain

solution

VOMS

will

reduce

costs

printers.

The

incineration

option allows higher VOM fountain solution,

 if needed

 for print

quality,

but

still

 results

 in effective control.

 Additionally,

there

are

other

factors

that

 support the economic reasonableness

the

rule

proposed

today.

The

levels

 control specified

 in the third First Notice

proposal are very close to the Agency rule that was analyzed in

the

EcIS.

The

EcIS

found

that,

even

statewide

basis,

the

cost of controls ranged from $808

 to $1,738 per ton, which was in

a

reasonable

cost

effectiveness

range.

 Revised and updated cost

estimates

for

 the incinerator control option were $300

to

$1,300.

 Revised cost estimates for the condenser/filter option

were

$170

$450

(Ex.

107).

The

rule

proposed

today

will

have

much

smaller

economic

impact

than

that

envisioned

the

EcIS,

First,

the

geographic

applicability

 Section 215,408(a)

is

limited

ten

counties

which

will

exclude

the

four

World

Color

Press

Inc.

facilities

from

add—on

control

 requirements.

 The EcIS

found that World Color Press

 Inc. would bear

 62 percent of the

statewide cost of control as a

result

add—on

control

costs.

Second, the approximately nine facilities and sixty—four presses

that will be controlled under 251.408(a)

 are already controlled

by either

 incinerators or condensers

 (P.C.

 82).

 These controls

are already in place because of smoke and odor regulations,

The

record

 indicates that the proposed control options are compatible

with control equipment now in use.

 This will further

 reduce the

cost of regulation from that estimated in the EcIS as initial

purchase and installation capital costs will not be incurred.

Calculating emissions and potential emission reductions

under

today’s

proposed

 rule involves a number of assumptions.

Because of the variability in emission factors

 arid the lack of

data on current VOM content of fountain solutions, especially

isopropanol substitutes, the emission

and

 emission reduction

figures

are

best

estimates.

such,

the

values

are

rounded

off

to two significant figures.

Based

 data supplied by the P11

for major stationary sources

 in non—attainment areas,

 it appears

that approximately 2400 tons/year of ink oils are used at nine

facilities that would be regulated under proposed Section

215.408(a)

and

(b)

 (P.C.

 82).

 Depending on the emission factor

77-340

—27--

used, this would result in an emission range of 1,700 tons/year

(at 0.70 emission factor)

 to 1,900 tons/year (at 0.80 emission

factor),

 P11 only provided data on IPA usage

 at these nine

facilities,

 As noted earlier,

 there

 are other VOM constituents

in fountain solutions other

 than isopropanol

 and, according

to

P11 witnesses, there

is

 a trend

 in the industry towards replacing

isopropanol with lower volatility VOM5.

 Consequently,

it

necessary to estimate fountain solution VOM.

 P11 estimated an

emission distribution ratio for the entire printing process of

60:40 at

current

 fountain solution VOM concentrations between 15—

25 percent (Ex.

 24(k)).

 In other words, at present ink and

isopropanol—based fountain solution usage,

 60 percent of the VOM

emissions are from the fountain solution and 40 percent of the

VOM emissions are attributable to the

 ink solvents

 (Ex. 24(k),

Ex. 71).

 Based on this ratio and the ink solvent data, the

estimated fountain solution VOM5 is

 2,800 tons/year at a 0.80

emission factor for

 ink solvents.

 Combining ink solvent and

fountain solution VOM emissions results

 in estimated total

emissions from the nine potentially regulated facilities of 4700

tons/year.

Emission reductions under proposed Section 2l5,408(a)(l),

 the incinerator option, are estimated by multiplying the removal

efficiency (RE) by the quantity of emissions,

 The RE for

fountain solutions is calculated by multiplying the fraction of

the fountain solution VOM presented

 to the incinerator by the

destruction efficiency of that incinerator.

 Emission factor

estimates for the fraction of fountain solution VOM presented to

the incinerator, via the dryer, range from 0,5 to 0.99.

Multiplying these figures by the 0.90 destruction efficiency of

the incinerator results in a RE range of 0.45

 to 0.89.

Multiplying these RE5 by the estimated fountain solution VOM

usage results

in

 a range

 of emission reductions of 1,300

tons/year to 3,500

 tons/year.

The

 RE for the ink solvents is calculated by multiplying the

emission factor by the destruction efficiency.

 The RE

 for the

ink solvents

 is 0.72 at 0.80 emission factor.

 Multiplying this

RE by the total

 ink solvent usage results

 in an ink solvent

emission reduction of 1,700

 tons/year.

 Combining the reductions

in fountain solution VOM and ink solvent emission results

 in a

range of potential emission reductions from 3,000 tons/year

to

4,200 tons/year.

 Actual emission reductions would vary within

this range.

Emission reductions under proposed Section 215.408(a) (2),

i.e. the fountain solution reformulation and condensation option,

are estimated somewhat differently than for 2l5.408(a)(l).

Section 215.408(a)(2)

 calls

for

 a reduction in fountain solution

VOM from current usage bevels of 15

to

 25 percent down to eight

percent.

 In this circumstance, emission reductions must be

estimated

 through the use of emission distribution ratios.

A

77-341

—28—

reduction of fountain solution VOM from 25

percent

eight

percent would change the emission distribution ratio of fountain

solution to ink solvents

 from 60:40 to 32:68.

 A reduction of

fountain

solution

VOM

from

percent

eight

percent

would

change

the

 emission distribution ratio of

 fountain solution to

ink

solvents

from

60:40

44:56,

These

ratios

can

used

combination

with

known

ink

solvent

usage

estimate

the

quantity

VOM5

the

fountain

solution

eight

percent

level.

While

it

 is impossible to determine what level fountain solution

 VOM5 are actually presently being

 used,

 a range of reductions can

be estimated,

 A reduction from 25 percent

to

 8 percent VOM in

the

 fountain solution would result

in

 a 68

reduction

VOM

usage.

 This corresponds to

 a 1,900

tons/year

reduction

fountain

solution

VOM.

reduction

from

percent

VOM

the

fountain

solution

would

result

reduction

VOM

usage.

This

 corresponds

to

 a 1300 tons/year reduction

fountain

solution

VOM.

Ink solvent emission reductions achievable through the use

condenser/filter

are

calculated

multiplying

the

quantity

emissions

presented

the

control

equipment

the

RE..

The

RE

for

the

condenser/filter

determined

multiplying

the

emission factor of 0.80 by the capture and removal efficiency of

 the condenser/filter, which is

 0.75.

 The RE

 is, thus,

 0,6.

The

then

multiplied

the

total

 ink solvent usage

 at the nine

facilities of 2,400 tons/year.

 This results

 in 1,400 tons/year

of ink solvent emission reductions

 in the condenser/filter.

Total

 emission reductions under Section 2l5.608(a)(2), which

includes

both

fountain

solution

VOM

reductions

and

reductions

from the condenser/filter option, range from

2,700

tons/year

3,300

tons/year.

There

are

eight

facilities

located

attainment

areas

(and

not

considered

part

 the Chicago urbanized area)

 that would be

subject

Section

 215,408(b),

 the fountain solution VOM

limitation of eight percent,

 Total organic emissions (fountain

solution

VOM

and

ink

solvents)

from

these

facilities

range

from

2,700

5200

tons/year

(Ex.

71).

Assuming

the

60:40

distribution between fountain solution VOM and ink solvents when

 traditional fountain solution

 is used results

total

fountain

solution VOM

emissions

1500

 3100 tons/year.

 A reduction of

fountain solution VON from 25 percent to eight percent results in

removal

1,100

 2100 tons/year.

 A reduction of fountain

solution

VOM

from

percent

eight

percent

results

removal

750

1,500

tons/year.

While

impossible

 determine

the actual present fountain solution VOM content,

 these figures

provide

 a reasonable estimated

 range of reduction.

The

estimates

potential

organic

material

emission

reductions under

 the two PACT alternatives, that involve the use

of add—on controls, demonstrate that the alternatives are roughly

comparable.

 Because the actual emission reductions at any one

77-342

—29—

facility

can

only

estimated,

 it is not possible

 to demonstrate

exact

equivalency

either

terms

reduced

emission

cost..

However, the potential emission reductions and costs do appear to

least

 a comparable range.

The

Board

believes

that

the

proposed

rule

represents

PACT,

Fountain

solution

VOM

reduction

through

reformulation,

required under Section 215.408(a) (2)

and

(b)

 are essentially no

cost

options

and,

fact,

will

save

printers

money

through

overall

reduction

isopropanol

and

isopropanol

substitutes,

The eight percent limit is considered technically feasible by the

P11.

 The add—on control options required under either 215.408(a)

(1)

 (afterburners)

 or 2l5,408(a)(2)

 (condenser/filter) are

clearly available control

 technology, as the record

 indicates

that such controls are already

 in place at the regulated

facilities.

 Costs for the afterburner option have been estimated

in the range of $300

—

 $1300 per ton of VOM removed.

 Costs for a

condenser/filter are estimated

 in the range of $170

—

 $450 per

ton of VOM removed (Ex.

 107).

 These costs are clearly within

a

reasonable range.

 The potential emission reductions from today’s

proposed rule are large when compared with many other PACT

industrial categories.

 The additional

 emission reductions that

will occur due to the attainment area fountain solution VOM

reduction are justified by the record,

 While this level of

control

 is not as stringent as the application of PACT in non—

attainment counties, the emission reductions are achieved at

essentially no cost.

 General background ambient HC and ozone

levels will be reduced.

 This will help maintain ozone attainment

throughout much of the state and also reduce the quantity of

ozone and ozone precursors available for atmospheric transport to

non—attainment areas.

 At least one major facility in Randolph

County

 is contiguous to the East St. Louis Metropolitan ozone

non—attainment region.

 Cost effective controls

 in such

circumstances are, therefore, prudent and justified.

Order

The following regulatory language

 is proposed for first

notice.

 The Clerk

 of the Board

 is directed to submit this

language

 to the Secretary of State for publication

 in the

Illinois Register.

TITLE

35:

 ENVIRONMENTAL PROTECTION

SUBTITLE B:

 AIR POLLUTION

CHAPTER

I:

 POLLUTION CONTROL BOARD

SUBCHAPTER c:

 EMISSION STANDARDS AND

LIMITATIONS FOR STATIONARY SOURCES

PART

215

ORGANIC MATERIAL EMISSION STANDARDS AND LIMITATIONS

SUBPART P:

 PRINTING

AND

PUBLISHING

77-343

—30—

Section

215.401

 Flexographic and Rotogravure Printing

215.402

 Exemptions

215.403

 Applicability of Subpart K

215,404

 Testing and Monitoring

215.405

 Compliance Dates and Geographical Areas

215.406

Alternative

Compliance

Plan

215.407

 Compliance Plan

215.408

 Heatset Web Offset Lithographic Printing

SUBPART P:

 PRINTING AND PUBLISHING

Section 215.401

 Flexographic and Rotogravure Printing

No owner or operator of a packaging rotogravure, publication

rotogravure or

 flexographic printing press subject to this rule

and employing solvent—containing

 ink may cause or allow the

operation of such press

 unless:

a)

 The volatile fraction of ink as it

 is applied

 to the

substrate contains 25 percent or less by volume of

organic solvent and

 75 percent or more by volume of

water; or

b)

 The volatile fraction of an ink as

it

 is applied

 to the

substrate,

 less water,

is

 percent or

 less by volume;

c)

 The owner

 or operator installs

and

 operates:

1)

 A carbon adsorption system which reduces the

volatile organic emissions from the capture system

by at least 90 percent by weight;

or

2)

 An afterburning system which oxidizes at least 90

percent of the captured nonmethane volatile organic

materials (measured as total combustible carbon)

to

carbon dioxide and

 water;

or

3)

 An alternative volatile organic material emission

reduction system demonstrated to have at least

a

90

percent overall reduction efficiency and approved

by the Agency;

and

d)

 A capture system

 is used

 in conjunction with any of the

emission control systems

 in subsection

(C).

 The design

and operation of the capture system must be consistent

with good engineering practice and shall provide,

in

combination with the control equipment,

 an overall

reduction

 in volatile organic material emissions of at

least:

77-344

—31--

1)

 75 percent where a publication rotogravure process

employed;

percent

the

maximum

reduction

achievable

using

 good engineering design where a packaging

rotogravure process

 is employed;

or

3)

 60 percent where a flexographic printing process is

employed.

Section 215.402

 Exemptions

The limitations of this Subpart shall not apply to any facility

whose aggregate uncontrolled rotogravure and/or flexographic

printing press emissions of volatile organic material are limited

operating

permit

conditions

 907 Mg (1000

 tons)

 per year or

less

 in the absence of air pollution control equipment or whose

actual emissions in the absence of air pollution control

equipment would be less than or equal

 to 907 Mg

(1000

tons)

per

year when averaged over the preceding three calendar

 years.

Section 215.403

 Applicability of Subpart K

Upon achieving compliance with this Subpart,

 the emission source

is not requrred to meet Subpart

K..

 Emission sources exempted

from this Subpart are subject

 to Subpart K.

 Rotogravure or

flexographic equipment used

 for both roll printing and paper

coating are subject

 to this Subpart.

Section 215.404

 Testing and Monitoring

a)

 Upon

 a reasonable request of the Agency,

 the owner or

operator of a volatile organic material source subject

 this Subpart shall

 at his own expense demonstrate

compliance

methods

procedures

approved

the

Agency.

b)

 A person planning

 to conduct a volatile organic material

emissions test to demonstrate compliance with this

Subpart

shall

notify

the

Agency

that

intent

not

less

than

days

before

the

planned

initiation

the

tests

so the Agency may observe

 the test,

Section 215,405

 Compliance Dates and Geographical Areas

a)

 Except as otherwise stated

 in subsection

 (b), every

owner

 or operator

 of an emission source subject

 to: t~s

77-345

—32--

6~pa~t

ehe4~

eomp~yw~tl’i4~ss~e~s

 at~~4me~e~s

~y Beeembe~~

Section

205.401

 shall comply with its standards and

limitations

December

31,

1983;

and

 Section 215.408

 shall comply with

its

standards

and

limitations by December 31,

 1987.

b)

 If an emission source subject

 to Section 215.401

 is not

located

one

the

 counties listed below

and

also

not located

 in any county contiguous thereto, the owner

or operator of the emission source shall comply with the

requirements of this Subpart no later than December 31,

1987:

Cook

Macoupin

DuPage

Madison

Kane

Monroe

Lake

 St. Clair

fBea~dNe~ei~These ee~iit4esare prepese~~e ~e

~es~na~e~

 as ne~a~ta~rimen~

 ~y ~i~e~SEPA +4~f~’ed~Re~

~3~&~T

~

 2~ ~992-)-)-

c)

 Notwithstanding subsection

 (b),

 if any county

is

designated

nonattainznent

the

USEPA

any

time

subsequent to the effective date of this Subpart, the

owner

operator

emission

 source located

 in that

county

any

county

contiguous

that

county

who

would

otherwise be subject to the compliance date

in

subsection

 (b) comply with the requirements of this

Subpart within one year from the date of redesignation

but

 case later than December 31,

 1987.

(Source: Amended at

Ill.

Reg.

 effective

___________)

Section 215,406

 Alternative Compliance Plan

The owner or operator of an emission source subject to this

Subpart may in lieu of compliance with Sections 215.405 and

215,407 demonstrate compliance through the use of a low solvent

ink program by taking

 the following actions:

a)

 Submit to the Agency a compliance plan,

 including a

compliance completion schedule, by December 31,

1983

which demonstrates:

1)

 Substantial emission reductions early in

the

compliance schedule;

77.346

—33—

2)

 Greater reductions

in

emissions

than

would

have

occurred without a low solvent ink program; and

3)

 Final compliance as expeditiously as possible but

 later

 than December

 31, 1987;

and

b)

 Certify to

 the Agency that:

1)

 A low solvent ink compliance strategy is not

technically available which would enable the

emission source

 to achieve compliance by the date

specified

Section

215.405;

and

2)

 An unreasonable economic burden would be incurred

if the owner or operator were required

to

demonstrate compliance by the date specified

in

Section

215.405;

and

Agree

install

one

the

control

alternatives

specified

Section

215.401(c)

June

31,

1986

the

specified

low—solvent

ink

strategy

fails

achieve

scheduled reductions by December 31,

 1985.

Section

215.407

Compliance

Plan

The

owner

operator

emission

source

subject

Section

2l5.405(a)(l)

shall

submit

the

Agency

compliance

plan,

pursuant

Ill.

Adm..

Code

201,

Subpart

including

project

completion

schedule

 where

applicable,

later

than

April

21,

1983.

The

owner

operator

emission

source

subject

Section

215.405(b)

shall

submit

 the Agency a

compliance

plan,

 including

 a project completion schedule

where

applicable,

later

than

December

31,

1986.

 The owner or operator of an emission source subject to

Section 215.405(c)

 shall

 submit

 a compliance plan,

including

protect

completion

schedule

within

days

after

the

date

redesignation,

but

case

later

than

December

31,

1986.

Unless

the

submitted

compliance

plan

schedule

disapproved by the Agency, the owner or operator of a

facility

emission

source

subject

the

rules

specified

 in subsections

 (a),

(b)

or

 (c) may operate the

emission

source

according

the

plan

and

schedule

submitted.

The

plan

and

schedule

shall

meet

the

requirements

Ill, Adm. Code 201, Subpart

H,

 including specific

interim dates as required

 in 35

 Ill,

 Adm..

 Code 201.242,

77-347

—34—

(Source: Amended at

 Ill.

 Reg.

 effective

____________)

Section 215.408

 Heatset Web Offset Lithographic Printing

a)

 No owner

 or operator of a heatset web offset

lithographic printing facility, located

 in Cook, DuPage,

Kane,_Lake, Maco~pin,Madison, Mcflenry, Monroe, St.

Clair or Will County, emitting over 100 tons/year of

organic material,

 in the absence of pollution control

equipment, may cause or allow the operation of

 a heatset

web offset press unless:

1)

 An incinerator system is installed and operated

that oxidizes at least 90 percent of the organic

materials (measured as total combustible carbon)

in

the dryer exhaust airstream to carbon dioxide and

water; or

2)

 The fountain solution contains no more than eight

(8) percent, by weight, of volatile organic

material and

 a condensation recovery system

is

installed and operated that removes at leat 75

percent of the non—isopropyl alcohol organic

materials from the dryer exhaust airstream.

owner

operator

 of a heatset web offset

lithographic

printing

 facility, located

in

 a county

other

 than Cook, DuPage, Kane, Lake, Macoupin, Madison,

Mcflenry, Monroe, St.

 Clair or Will County, emitting over

100

tons/year

 organic material,

 in the absence of

pollution control equipment,

 may cause or

 allow the

operation of

 a heatset web offset press unless the

founta_n solution contains no more than eight

(8)

per-

cent,

 by weight,

 of volatile organic material.

(Source: Added

at

 Ill. Peg.

 effective

___________)

IT IS SO ORDERED

I, Dorothy M.

 Gunn, Clerk

 of the Illinois Pollution Control

Board, hereby certify that the above Proposed Rule,

 First Notice

Opinion

and

Ordeç

was

adopted

the

...3ot7~

day

_______________________,

1987,

vote

Dorothy

 M.. Gum,

 Clerk

Illinois Pollution Control Board

77~348