Guidelines for Oxygenated Gasoline Credit Programs under
    Section 211 (m) of the Clean Air Act as Amended
                     U.S. ENVIRONMENTAL PROTECTION AGENCY
                                        Office of Mobile Sources
                             Field Operations and Support Division

-------
Guidelines for Oxygenated Gasoline Credit Programs under Section 21 1 (m) of the
Clean Air Act as Amended                           -

SUMMARY:        Section 21 1 (m) of the Clean Air Act as Amended by the Clean Air
Act Amendments of 1990 ('the Act") requires that various states submit revisions to
their State Implementation Plans (SIPs), and implement oxygenated gasoline
programs. This requirement applies to all states with carbon monoxide (CO)
nonattainment areas with design values of 9.5 parts per million or more based
generally on 1988 and 1989 data.  The oxygenated gasoline program must require
gasoline in the specified control areas to contain at least 2.7% oxygen by weight
during that portion of the year in which the areas are prone to high ambient
concentrations of carbon monoxide.

     Section 21 1 (m)(5) of the Act  requires that EPA promulgate guidelines for state
credit programs, allowing the  use  of marketable oxygen credits for  gasolines with a
higher oxygen content than required to offset the sale or use of gasolines with a lower
oxygen content than required.

      These guidelines are for such oxygenated gasoline  credit programs.

ADDRESSES:      Materials  relevant to these guidelines have been placed  in Docket
A-91-04 by EPA.  In addition,  EPA has engaged in the  Regulatory Negotiation process
to assist in developing these guidelines.  A separate docket has been set up for the
Regulatory Negotiation, Docket A-91-17.  These dockets are located in the Air Docket
Section (LE-131), U.S. Environmental  Protection Agency, 401 M Street, S.W.,
Washington, D.C. 20460, in room  M-1500 of Waterside Mall, and may be inspected
from 8:30 a.m. to 12:00 noon and from 1:30 p.m. to 3:30 p.m., Monday through
Friday. A reasonable fee may be  charged for copying  docket material.
FOR FURTHER INFORMATION CONTACT:  Alfonse Mannato, (202) 26^9050

SUPPLEMENTARY INFORMATION:

I.  Introduction

      .These guidelines are for oxygenated gasoline credit programs required under
section 21 t(m) (5) of the Act.  The remainder of this preamble is divided into three
parts. SedSon II  provides the background for this action, with respect to chronology
and the broad issues involved. Section III presents EPA's action and rationale.
Section IV summarizes and addresses comments received in response to the Notice
of Proposed Guidance1 published on July 9, 1991 and the Supplemental Notice of
   1 56 FR 31148 (July 9, 1992).

-------
 Proposed Guidance2 published on February 5,1992.

 II. Background

      Motor vehicles are significant contributors of carbon monoxide emissions.  An
 important measure to reduce these emissions is the use of oxygenates in motor
 vehicles' gasoline.  By adding oxygenates to gasoline, exhaust emissions of carbon
 monoxide are reduced.

      Section 211 (m) of the Act requires that states with carbon monoxide
 nonattainment areas with design values of 9.5 parts per million or more, based on
 data for the two year period of 1988 and  1989,3 submit revisions to their State
 Implementation Plans (SIPs). Although the Act does not specify a due date for these
 SIP revisions, the Agency is interested in  setting such a date in  order to encourage
 consistency across the nation in implementing the oxygenated gasoline programs.

      Out of three possible due dates proposed by the Agency in its February 5, 1992
 Supplemental Notice of Proposed Guidelines,4 today's guidelines specify November
 15, 1992 as the due date for the submittal of section 211(m) oxygenated gasoline SIP
 revisions. Many comments received by the Agency on this issue expressed the view
 that the states would have difficulty complying with an earlier submittal "target date"
 because of the complexities involved and the tight timelines associated  with
 implementing an oxygenated gasoline program. In addition, this date is consistent
 with other CO SIP call provisions which are specified in section  187 of the Act.  Also, a
 date earlier than  November, 1992 would likely require notice and comment rulemaking.

      The SIP revisions for the required CO nonattainment areas must establish
 oxygenated gasoline programs requiring at least 2.7% oxygen by weight,  except that
 states may adopt credit programs  such that gasoline with a higher oxygen content
 than required can offset the sale or use of gasoline with a lower oxygen content than
 required.  The oxygen content requirement is subject  to a testing tolerance to be
 established by the Administrator.  The oxygen content requirement applies during the
 portion of the year in which the areas are prone to high ambient concentrations of
 carbon monoxide.
    2 57 FR 4414 (February 5, 1992).

    3 The Agency has determined that the 1988 and 1989 data from several areas is
inadequate to properly characterize the ambient concentrations of CO.  Therefore, for
these areas - Boston, Cleveland and Washington, D.C. - the older, more representative
data has been used.

    4 57 FR 4414 (February 5, 1992).

-------
      Under the Act, the length of these control periods is to be established by the
Administrator and shall not be less than four months in length unless a State can
demonstrate, based on meteorological conditions, that a reduced period for any
individual control area will assure that there will be no carbon monoxide exceedances
outside of such period. These requirements are to cover all gasoline sold or
dispensed in the larger of the Consolidated Metropolitan Statistical Area (CMSA) or the
Metropolitan Statistical Area (MSA) in which the nonattainment area is located.
Guidance on the establishment of control periods appears in additional guidelines
published separately.5

      Section 211 (m) (5)  of the Act requires that the Administrator promulgate
guidelines allowing for the use of marketable oxygen credits from gasolines with a
higher oxygen content than required to offset the sale or use of gasolines with a lower .
oxygen content than required. Under that  subsection, oxygen credits may not be
transferred between control areas but instead may be used only in the area in which
they were created.

      These guidelines are for state oxygenated gasoline credit programs. The
guidelines include an enforcement scheme with responsibilities and liabilities of various
parties involved in the oxygenated gasoline industry.

      This document presents EPA's guidance to states regarding credit programs to
be employed in oxygenated gasoline programs under section 211 (m) of the Act.  This
guidance is a general statement of policy.  It does not establish a binding norm and is
not finally determinative of the issues addressed.  Agency  decisions in any particular
case will be made applying the law, applicable regulations and guidelines on the basis
of specific facts and actual action.

      To expedite Agency decisions in .particular cases, a state submitting a SIP
revision which includes an oxygenated gasoline credit program should identify all
areas where the state program differs from these guidelines  and provide as detailed an
explanation as possible for these differences.  For example, this explanation could
include, but need not be limited to, an explanation of any circumstances unique to the
state or localities involved and a demonstration of whether the state's proposed
program would be at least as effective as the program presented in this guidance.

      EPA is aware that the gasoline production and distribution industry extends to
all areas of the country, crossing state borders in an intricate, nationwide web of
   5 EPA is publishing final guidelines for the three oxygenated fuels-related notices
which were published at 56 FR 31148, 31151 and 31154 on July 9, 1991. Two of
these notices were published as supplemental notices at 57 FR 4408 and 4413 on
February 5, 1992.

-------
commerce. At the same time, the oxygenated gasoline programs required by the Act
are centered around a limited number of carbon monoxide nonattainment areas and
their surrounding CMSAs or MSAs.  State-based oxygenated gasoline credit programs
should be structured in a way that assures their successful implementation, to the
greatest extent possible, within the limits of state authority over a nationwide
production and marketing structure.  Coordination among states is specifically
addressed in sections 102 and 187(e)(1) of the Act. EPA believes that these
provisions reflect Congress's concern that state programs applicable to multi-state
nonattainment areas be coordinated, with the Agency's help.

      EPA will attempt to minimize problems associated with multi-state MSAs and
CMSAs.  The Agency is committed to providing technical support to the states in
implementing thes,e oxygenated gasoline guidelines. Theses guidelines should help
ensure program consistency in multi-state program areas.  The Agency plans to
provide technical support such as standardized training materials, audit forms,  industry
report forms, and database software to state officials.  Also, the Agency will encourage
cooperative activities by the states in an attempt to coordinate the implementation of
these multi-state programs.

Regulatory Negotiation
      EPA used the Regulatory Negotiation process to aid in the development of
these guidelines.  This process was initiated on February 8, 1991 when EPA
announced its intent to form an Advisory Committee to negotiate certain guidelines
and proposed regulations implementing the clean fuels provisions of Section 211 (k)
and (m) of the Act.6 A public meeting was held on February 21-22,1991 in
Washington, D.C. and after considering comments submitted in response to the notice
and the results of that public meeting, an Advisory Committee was established on
March 13,19917  On August 16,1991 an Agreement in Principle was signed by
members of the Advisory Committee.  That Agreement embodies the consensus orr
the part of the Advisory Committee members on basic elements of the oxygenated
gasoline and reformulated gasoline programs.  A copy of the Agreement has been
placed in .the public docket.  The above-referenced notices contain a more detailed
discussion of the issues referred to the Advisory Committee, as well as information on
the requirements of the regulatory negotiation process.

Summary  of the Guidelines
    These guidelines are to be employed in state oxygenated gasoline programs in
which gasoline containing more oxygen than the minimum 2.7% by weight than is
required would generate marketable credits.
   6 56 FR 5167 (February 8, 1991).

   7 56 FR 10522 (March 13, 1991).

-------
      The credit program guidelines here presented by EPA are designed to ensure
that all gasoline sold or dispensed in the control area, on the average, meets or
exceeds the minimum oxygen content required under Section 211 (m).

      Gasoline is typically sold or dispensed from gasoline terminals into trucks for
shipment to retail stations, or transferred in bulk to other terminals.  Under the
Agency's averaging program guidelines, averaging at the terminal level is required. In
addition, oxygenate or gasoline blending conducted in trucks at the terminal or at
remote locations is to be included in such averaging.  This scenario should
encompass all retail gasoline in a control area and should thus result in all such
gasoline meeting the required oxygen content on the average. Taking advantage of
the terminals' central position  in the gasoline distribution system  should maximize the
credit program's success while minimizing its burdens, both on the regulated
community and the governmental bodies involved.

      The party responsible for complying with the minimum 2.7% oxygen by weight
standard on the average, over the designated averaging period,  must be specifically   -
identified.  This party will be designated the Control Area Responsible Party (CAR).
The responsibilities of the CAR are discussed more fully below.

      At the terminals the CAR would be the person who owns the gasoline sold or
dispensed from a control area terminal into a truck.8 Selling or dispensing gasoline
from a terminal into trucks is commonly referred to as "breaking  bulk."  Parties who
own or operate terminals but who do not own or sell gasoline are not CARs. Also,
persons downstream from a control area terminal who blend oxygenates into gasoline
or who otherwise change the  oxygen content of gasoline intended for use in any
control area are also CARs, but are called Blender CARs. Blender CARs and CARs
are hereinafter collectively referred to as CARs.  Terminal owners, whether or not they
are CARs, must provide CARs using the terminal with the volume and oxygen content
of the gasoline delivered to or received from each CAR.

      The volume and oxygen content of all gasoline entering into a terminal must be
provided to the CAR.  Based on this and other information the CAR must keep a
running weighted average of the gasoline it transfers into each control area.9
Gasoline that is transferred in  bulk becomes the responsibility of the CAR to whom it is
transferred.  Gasoline transferred by a CAR to another CAR is therefore removed from
   8 Control area terminals would be those terminals at which gasoline intended for
use in any control area is sold or dispensed into trucks. The terminal itself need not
be physically located in the control area.

   9 Section 211 (m)(5) of the Act requires that an averaging program be conducted
separately for each control area.

-------
the averaging calculations of the CAR who transferred the gasoline.  At the end of the
averaging period, the average oxygen content of all gasoline the CAR distributed to
trucks destined for each separate control area is calculated separately.  In each
control area, if the average oxygen content is greater than or equal to the required
minimum, then compliance has  been demonstrated.  Credits are created if the avefage
is greater than the required minimum. If the average oxygen content is less than the
required 2.7% by weight minimum then credits are needed to meet the  compliance
average.

       The averaging program presented in these guidelines is similar to the type of
program used by EPA in the lead phasedown gasoline program. To comply with the
oxygenated gasoline program CARs must, at a minimum, achieve the sales-weighted
average oxygen content over a  specified time period, called the averaging period.
This can be done either by always selling each gallon of fuel with an oxygen  content at
or above the requisite oxygen content, or by. adjusting the quantities and types of fuel
sold over the averaging period either directly or by obtaining credits from  another
regulated party within the control period to attain at least the requisite oxygen content
on an averaged  basis.

      There is no intended prohibition or limitation on the ability of third party brokers
to facilitate the purchase and  sale of credits.  However, while persons other than CARs
may act as brokers, only CARs  may own credits. Since brokers may not be as
established in the industry as CARs, they may have a reduced sense of responsibility
for the program  requirements. Also, credits may be transferred to the extent such a
transfer would not result in any transferor having a negative credit balance at the
conclusion of any averaging period. Any credits transferred in violation of this
limitation are improperly created credits which may not be used regardless of the
transferee's good faith. Where any credit transferor has in its balance both credits
which were properly created and those which are improperly created, the properly
created credits should be applied first to the transfers before the transferor may apply
any credits to achieve its own compliance.

      Although not strictly necessary to achieve the desired air quality  results or to
comply with the  requirement of Section 211(m), an averaging program has a number
of benefits. The principal advantage of this program design is that it entails less
regulatory intrusion into the marketplace than traditional command and control
approaches.  It thus retains a high degree of marketing flexibility and competition
among blending agents. The advantageous aspects of this approach can be further
enhanced by allowing suppliers  to trade oxygen credits among themselves, with
suppliers of relatively low-oxygen fuels able to purchase such credits from suppliers of
relatively high-oxygen fuel  within a control area.

       Furthermore, when  compared to an oxygenated gasoline program  requiring
oxygen content compliance on a per-gallon basis, a program incorporating an oxygen

-------
averaging provision should prove to be less costly to implement in 1992.  This is due
to the fact that averaging programs will allow the supply of oxygenates, which some
parties have suggested to be limited for the first control season beginning in 1992, to
be used in a flexible and hence more  efficient manner.  Therefore, EPA strongly
recommends that states adopt averaging programs consistent with these guidelines.

      EPA and various parties have raised concerns about the possibility that in the
context of an averaging program, gasoline which does not conform to the
requirements for oxygenated gasoline may  be sold or dispensed within a control area
without  being detected since there would be no per-gallon standard against which to
test the gasoline. Two options were considered to address this potential problem.
The first option would have required the use of a marker.  Under that option, gasoline
whjch was not destined for use in a control area would  contain a marker  and it would
be a prohibited act for parties downstream  of the terminal in the  distribution system to
sell or dispense such marked gasoline in a control area. The second option was-to
establish a minimum oxygen content requirement for  all gasoline sold or dispensed
within the control area.

      EPA's guidelines specify a minimum  of 2.0% oxygen by weight in all gasoline
offered for sale, sold or dispensed by a CAR for use in  the control areas  during the
control period. This requirement would also apply to all parties.downstream of the
CAR.  The same minimum requirement would apply for  all gasoline sold or dispensed
to the ultimate consumer in the control area during the  applicable control periods. The
only exception to this requirement would be for gasoline sold or  dispensed from one
CAR directly to another CAR.  Also, this requirement does not prevent the storage of
nonoxygenated gasoline which is either intended for use in blending with  ethanol or
intended for use  outside of control areas.  EPA requests that states monitor the
availability of and demand for a variety of oxygenates, and should take appropriate
steps necessary  to reasonably assure the availability of  these various oxygenates in
the marketplace.

      In these guidelines, CARs are required to register with the state, and to provide
reports on each  averaging period.  Under the guidelines, each CAR which has
registered to engage in credit averaging and trading shall, as an additional part  of the
reporting requirements, submit an attestation engagement to the state. The attestation
engagement may be conducted  by either an internal auditor employed by the CAR or
a certified public accountant (CPA). The attestation engagement shall be based upon
the agreetf-upon procedures contained in these guidelines, and shall focus on the
underlying documentation that forms the basis of the reporting and recordkeeping
requirements under these guidelines.  This  attestation report will be made at the
conclusion of each annual control period.  An illustrative CPA report is included as
Appendix 0.

      These credit program guidelines provide that credits are created on the basis of

                                       8

-------
the oxygen content of the oxygenated gasoline sold or dispensed in a particular
control area, that credits are to be used to demonstrate compliance only within the
same control area in which they were earned, and that credits may only be used
during the averaging period in which they were created.

III.  Action

Sale of Only Oxygenated Gasoline in a Control Area
      The guidelines specify that each state establish a minimum 2.0% oxygen
requirement by weight in all gasoline sold or dispensed by a CAR for use in the
contrpl areas during the control period.  This requirement would also apply to all
parties downstream of the CAR. The same minimum requirement would apply for all
gasoline sold or dispensed to the ultimate consumer in the control area during the
control period. The only exception to this requirement would be for gasoline sold or
dispensed from one CAR directly to another CAR. This requirement would apply  in
addition to the other requirements  for averaging programs.

      There are several reasons for using a minimum content requirement as an
adjunct to an averaging program.  First, in each control area there will be less potential
variation in the effect of the program on the ambient air quality level on any given day,
because of the 2.0% oxygen by weight minimum.  Second, enforcement of the
program will be somewhat simplified in that state enforcement personnel could readily
take samples for comparison to the required minimum.10  Finally, there will be less
potential for consumer confusion concerning the amount of oxygen being marketed.

      EPA recommends that the states implement a five-day leadtime requirement.
The data suggests  that a leadtime  of five days will ensure that most, if not all, retail
stations will be able to dispense gasoline on the first day of every control period that
contains the 2.0% oxygen by weight minimum content requirement.  A longer period is
not necessarily supported by the data and may cause a significant reduction in the
supply of oxygenates available for  the oxygenated gasoline programs during the
control periods throughout the country, especially in the first year of the program.
EPA recommends that each state become familiar with the logistics associated with
their respective CMSAs and MSAs in order to ensure that a five-day leadtime is
adequate.

Length of Averaging Period
      The following averaging periods shall apply to all control areas: For any area
with a control period of five months or less, the averaging period shall be equal to the
    10 Sampling for the minimum will not obviate the need for states to audit CARs to
ensure the averaged standard of 2.7% is being met. Guidelines for state audits will be
covered in a separate implementation document to be issued later this year.

-------
control period, and for areas with control periods of six months or longer, the
averaging period shall be three months in length. In addition to these averaging
periods, EPA recommends use of a 30-day reconciliation period following each
averaging period, during which time CARs may purchase or sell credits for use in
connection with the averaging period which has just concluded. Because credits may
not be "banked," or carried from one control period to another, this reconciliation
period will allow maximum use of credits  by each CAR during each control period.

Banking Credits
      Some parties have suggested that the banking of credits from one averaging
period to another should be allowed as a means of permitting further flexibility to the
industry.  In light of the averaging periods that have been established, the minimum
requirement of 2.0% oxygen in every gallon of gasoline and the 30-day reconciliation
period at the end of each averaging period, the Agency believes sufficient flexibility is
available, is concerned that "carryover" of credits could potentially  present air quality
problems, and therefore is not including a banking provision in these guidelines.

Blendstock/Export/Storaoe Issues
    The sale or distribution of nonoxygenated gasoline by any person for use in any
control area is prohibited under these guidelines unless (a) such gasoline is
segregated from oxygenated gasoline, (b) the  documents which accompany such
gasoline are clearly marked as "nonoxygenated gasoline, not for sale to ultimate
consumer in a control area," and (c) the nonoxygenated gasoline is in fact not sold or
dispensed to ultimate consumers, during the control period, in the control area.
Gasoline  intended for sale to the ultimate consumer in a control area during the
control period must contain the required 2.0% minimum oxygen content to avoid
enforcement action at any point from the CAR to the retailer or the wholesale
purchaser-consumer.

   In classifying product, however, some concern has been expressed about.
blendstock, gasoline which is  destined for export, and gasoline in  storage. These are
petroleum products that are not standard oxygenated gasoline and would not contain
the required 2.0% oxygen content, but which might have a legitimate presence within a
control area.

    As a matter'of enforcement policy, EPA expects that a state would not hold a party
liable for the possession or transfer of nonoxygenated product downstream of the
refiner or importer facility which may arguably  meet the regulatory definition of gasoline
If the following requirements are met:

      1.  The product is clearly labeled as "blendstock/expprt/storage" and the
evidence supports this classification;
      2.  The Iccompanying  documents clearly state that the product does not
comply with the oxygenated gasoline  requirements;

                                       10

-------
       3.  Some aspect of the product's quality supports the party's claim that the
product was intended to be further blended before being sold, supplied, etc. as
finished product;
       4.  The seller, supplier, or transporter of the product has obtained a written
certification or notice on shipping documents from the buyer/recipient of the product
that the buyer/recipient understands that the product is not intended for sale or
distribution as finished gasoline in a control area unless or until (a)  it is blended to
meet the oxygenated gasoline regulatory requirements; (b) the buyer/recipient
receives equivalent certification from  a subsequent buyer or obtains a written
certification that the gasoline  will not  be sold or dispensed  for use within a control
area; and
       5.  The party has no knowledge or reason to believe that the product will not be
further blended to comply with the regulatory standards before being  sold, supplied or
transported as finished product, or that it would be sold or dispensed without further
blending within a control area.

Registration
    At least 30 days before the beginning of any control period, any person who sells
or dispenses gasoline  obtained from a control  area terminal, known as a CAR, would
be required to register with the state. At the time of registration, every CAR would be
required to declare its  intention to comply with these guidelines through oxygen credit
unit averaging or per-gallon minimum standards.  Any party which decides it would like
to participate in  a control area as an averaging CAR, either prior to or after the start of
the control period, may do so after submitting  a  registration application to the state,
and receiving the state's approval. States are  encouraged to process the applications,
especially those received in mid-season as quickly or possible. When registered by a
state, every CAR would receive a CAR identification number, with averaging CARs
authorized to conduct  oxygen credit  unit trades.  This  registration would be on a form
provided by the state,  and would contain basic information provided by the
owner/operator on the day-to-day operation of the terminal or blending facility from"
which the CAR operates. The state would have the flexibility to request additional
information that it deems appropriate. A valid  registration would be a pre-condition for
operating as a CAR. From the time any such  information becomes inaccurate, the
CAR would have 30 days in which to provide an update. The 30-day time period
would allow the states sufficient time for the  review of applications,  while still allowing
the CARs the flexibility to join state averaging programs during a control season.

Specific Responsibilities/Liabilities of Regulated Parties
      The oxygenated gasoline credit program guidelines  impose responsibilities on
parties in the gasoline  industry which fall generally into four categories:

             Persons who  produce or import gasoline (refiners and-importers)
      are responsible for assuring that gasoline is tested and that the
      documentation that accompanies the  gasoline accurately reflects the

                                        11

-------
      oxygen content.  Liability for violations of these requirements is for the
      refiner or importer only.

             Persons who transport, store or sell gasoline (refiners, importers,
      blenders, distributors, resellers, retailers, wholesale purchaser-consumers and )
      have various responsibilities associated with assuring that only oxygenated
      gasoline is sold or dispensed for use in control areas. Persons who transport,
      store, or sell gasoline downstream from the CAR are responsible for assuring
      that gasoline intended for sale to retailers or wholesale-purchaser consumers
      within a control area meets the 2.0% required minimum oxygen specification.
      Persons who transport, store, or sell gasoline at the terminal or upstream from
      the terminal are responsible for assuring that the oxygenate content of all.
      gasoline intended for use in a control area, as stated on the accompanying
      paperwork, is accurate. These persons are also responsible for assuring that
      all nonoxygenated gasoline sold into a control area for use as a blendstock is
      sold only to CARs duly registered with the state.  Liability for violations of these
      requirements is for the facility where the violation is found, and for all  persons
      upstream from that facility, except in the case of violations associated with the
      minimum requirement, which stop at the terminal.

            Terminal owners and operators are responsible for assuring that the
      oxygen content of the gasoline they receive, handle or dispense is  accurate.
      CARs are responsible for assuring that gasoline intended for use within a
      control area, during the control period, meets the 2.0% required minimum
      oxygen specification; for obtaining the information necessary to assure, if
      possible, that oxygenated gasoline once accounted for is in fact sold  or
      dispensed  in the  proper control area; for properly accounting for credits
      generated, transferred or received; for assuring that the oxygenated gasoline
      standard is met on the average for each averaging period in each relevant
      control area; and for submitting reports evidencing compliance.

            Retailers and wholesale purchaser-consumers are responsible  for
      assuring that gasoline intended for sale during the control period contain at
      least 2.0% oxygen, by weight.

      The term "responsible for assuring" as used above is not meant to imply that a
party guarantee compliance at a point downstream from It in the gasoline distribution
network,  fn fact,  elements of various defenses that would  be available to regulated
parties are discussed below.

      With respect to those regulatory responsibilities where potential liability exists for
parties upstream  from the facility found in violation, the  guidelines  include liability for
the operator of the facility in violation and presumptive liability for upstream parties.
Under this approach, defenses would be available for each party with presumptive

                                       12

-------
liability.  This is the scheme which is followed under the federal gasoline lead
contamination, volatility, and diesel fuel sulfur content regulations.11

      EPA believes that the principal advantage of the presumptive liability approach
is that it allows  identification of the person who caused the violation.  EPA is
concerned that nonoxygenated gasoline could be mixed with oxygenated gasoline by
any person in the gasoline distribution network, and that it would be difficult or
impossible for the state to identify the person responsible for causing this violation.  In
order to address this difficulty, those persons who actually handle the gasoline, who
are in the best position to identify the cause of any violation, must have an incentive to
be forthcoming in providing accurate compliance information. EPA believes that a
presumptive liability scheme is the most appropriate method of addressing this
concern.  This is a scheme which is familiar both to EPA and to industry, and makes
the most efficient use of state resources.

      Under a  scheme of presumptive liability there are a number of defenses
available to all parties in the oxygenated gasoline distribution network.  In any case in
which a refiner, importer, CAR or blender CAR, distributor, reseller or carrier would be
liable for a violation detected downstream of its own facility, any party can construct a
defense by demonstrating three things:

      1)    That the violation was not caused by the regulated party or its
employees or agents;

      2)    That the proper documentation accompanied the product and that the
documents contain accurate information regarding the oxygen content of the product
in question; and                                                 .

      3)    In the case of a refiner, importer, CAR or blender CAR,  distributor, or
reseller, that the regulated party conducts an acceptable sampling and testing
program to assure the  quality of the product being handled on a regular basis.

      The idea of a quality assurance program is not unique to these oxygenated
gasoline  guidelines; the Agency has employed such defense schemes in other fuel
programs such as  lead phasedown. The Agency feels that the easiest way for a
regulated party to prove that it is not liable for violations is through a random  sampling
and testing program; however, this type of testing is not required for all parties in the
gasoline  distribution network. Because by definition they never take ownership of the
gasoline  they handle, carriers will be held to a different standard for liability defense.
For refiners, importers, CARs and blender CARs, distributors and resellers, for a
quality assurance program to be acceptable to the Agency, the regulated party would
    11 See 40 CFR 80.23, 80.27 and 80.29.

                                       13

-------
 need to demonstrate that it either conducts the program itself, or commissions an
 independent testing service to conduct systematic periodic testing of a percentage of
 the total product handled by that regulated party.  The program should be designed to
 provide an accurate representation of a sample of products handled by the regulated
 party.                                                                    .

      The design of these quality assurance programs would be flexible, a matter to
 be considered by each individual regulated party and the state enforcing its
 compliance.

      For the forgoing reasons, EPA has designed a liability scheme for the
 oxygenated gasoline credit program guidance based upon presumptive liability. EPA
 believes such an approach will be the most effective and equitable method of placing
 liability Upon the party or parties responsible for causing a violation.

 The Control Area Responsible Party • The Control  Area Responsible Parties (CARs)
 are those parties subject to the average oxygen content standard.  To account for
 oxygenated gasoline credits units, the CAR must know the specific oxygen content of
 each batch of oxygenated gasoline it has delivered to a control area to be offered for
 sale or dispensed by a retailer or a wholesale purchaser-consumer during the control
 period.

      The guidelines direct that there be two potential responsible parties.  The first
would be the person who owns gasoline which is sold or dispensed over the rack
from a control area terminal, or the CAR.  A control area terminal is a facility which is
capable of receiving gasoline  in bulk, i.e. by pipeline or barge, and/or at which
gasoline is altered either in quantity or quality. Gasoline which is intended for use in
any control area is sold or dispensed into trucks at these control area terminals.  The
second potential responsible party would be the person who  owns oxygenated
gasoline which is sold or dispensed from a control, area oxygenate blending facility, or
the Blender CAR. A control area oxygenate blending facility is any facility or truck at
which the oxygen content by weight of gasoline intended for use in any control area is
altered in any manner other than combining two or more fuels complying with the
2.0% oxygen by weight minimum requirement. The quality or quantity of gasoline is
not altered in any other manner, except through the addition of deposit-control
additives.  All CARs and Blender CARs will be required to  register with the state before
being allowed to buy or sell oxygenated gasoline or oxygen credit units.

      Questions have been raised  as to the regulatory status of various parties such
as a truck blender who blends raffinates or who adds gasoline to an oxygenate blend.
The definitions have been expanded to include these various  scenarios. Such
"blenders" would be considered a CAR if he sells or dispenses blended gasoline in a
control area during a control period since he is altering both the quantity and the
quality of gasoline.


                                       14

-------
      At gasoline terminals which sell or dispense gasoline for use in a control area,
the owner of the gasoline which is sold or dispensed is the CAR.  The CAR must know
the oxygen content of the gasoline it is dispensing or selling in order to account for
the credits or debits generated by that gasoline and to ensure that every gallon
complies v/ith the minimum oxygenate requirement of 2.0% oxygen content by weight.
The CAR would know this information through receipt of transfer documents from
upstream  parties, through its own testing as part of a quality assurance program, or
by receipt of information on the mathematically calculated oxygen content from the
terminal operator in charge of the terminal from which a CAR'S gasoline was sold. It is
the CAR'S responsibility, at the close of every averaging period, to demonstrate
compliance with the average 2.7% oxygen content by weight for the total volume of all
gasoline sold or dispensed  into any one control area over the course of the entire
averaging period.

      CARs which have chosen to comply with these guidelines through a per-gallon
standard would have less of a burden to prove compliance with an oxygenated
gasoline program.

      When any blending of oxygenates occurs at the terminal or at another location
downstream from the terminal, the responsible party is the Blender CAR.  Owners of
gasoline who are not registered CARs are permitted to sell gasoline only outside of
control areas, or to sell to registered CARs and Blender CARs.  Once a Blender CAR
has obtained the gasoline, it may add oxygenates to it in order to comply with the
average oxygen content standard and the minimum per-gallon oxygenate requirement
of 2.0% oxygen by weight.  It  is the Blender CAR'S responsibility, at the close  of every
averaging period, to demonstrate compliance with the average 2.7% oxygen content
by weight for the total volume of all gasoline it sold or dispensed into any one control
area over the course of the entire averaging period.

      CARs and Blender CARs have the same responsibilities and liabilities.. Their
responsibilities consist generally of accounting for all oxygen content associated with
the oxygenated gasoline which is dispensed into trucks for delivery into any control
area during a control period, to ensure that every gallon sold or dispensed for use in
the control area meets the 2.0% minimum oxygen requirement,  and for submitting
reports to the state at the conclusion of each averaging period showing average
oxygen gasoline standards were achieved.

      EPA's guidelines provide that  CARs shall, as part of the reporting requirements,
engage either an internal auditor or an independent certified public accountant (CPA)
or firm of CPAs to perform an agreed-upon procedures attestation engagement of the
underlying documentation that forms the basis of the reports which are submitted to
the state in accordance with the requirements of the guidelines  at the conclusion of
each annual control period. This requirement is discussed more fully below.
                                       15

-------
      The averaging responsibility is located at the gasoline terminals and at blender
facilities because, as previously described, they represent the last centralized point in
the gasoline distribution network before gasoline is transported by truck to a wide
variety of retail locations in the control area.  EPA expects that compliance with the
averaging requirement at terminals and  blending facilities will lead to compliance, on
average, by all gasoline dispensed to ultimate consumers in the control area. The
centralized nature of the CAR facilities will also allow the averaging requirement to
apply to a manageable number of identifiable parties, facilitating implementation and
enforcement. Because the 2.7% oxygen by weight requirement is an average to be
applied over an entire control area, If a CAR supplies a single control area from more
than one terminal, the CAR may combine volumes sold from the respective terminals
to satisfy the average oxygen requirement.

      It is EPA's opinion that the owner of the gasoline is in the best position to
exercise control over the oxygen content of the gasoline. However,  EPA
acknowledges that there are various situations where the responsibilities of a CAR may
be shifted to another CAR which is in a  better position to assume such responsibilities.

      A CAR would be able  to contractually assign its responsibilities under these
guidelines to another party who is also a registered CAR by contractually transferring
ownership of the gasoline. Thus, a party which owns gasoline but which wishes to
reassign its CAR responsibilities (i.e. recordkeeping, reporting, averaging,  etc.) to
another party better-suited to act as the CAR for the gasoline in question (such as a
partner in an exchange agreement) could do so.  The assigning CAR would have a
zero volume of gasoline to report.

      In addition, these guidelines allow terminal owners and operators to act on
behalf of a CAR by accepting gasoline into the terminal, but not to allow its
introduction into commerce unless the proper documentation  accompanies it
containing information such as oxygen content and volume, or until testing using
approved methods has been done to establish the oxygen content.  This is designed
to assure that the information needed to conduct averaging is available to the CAR.
The terminal owner or operator would also be responsible for conducting  a quality
assurance program to verify  the accuracy of such information.

      Compliance in the oxygenated gasoline program for  CARs is based upon the
oxygenated gasoline dispensed into trucks or barges for transport into control areas,
plus or minus any credit transfers, and excluding the oxygenated gasoline transferred
outside the control area in bulk or to another registered CAR in any  control area.
Separate compliance determinations must be calculated for every control area served
by a CAR, regardless of the  number of terminal facilities owned by that CAR which
serve the same area.
            -*i  '
      The following is an example of a  compliance calculation for a  CAR.


                              .16

-------
      On day one of the compliance period the CAR received 100,000 gallons of
oxygenated gasoline, containing 3.0 percent by weight oxygen.  The credit status of
this batch of gasoline is calculated as follows:

Actual Oxygen Content Units = weight percent X gallons

      3.0 X 100,000 = 300,000 actual oxygen content units

      The CAR received a total of three other shipments of oxygenated gasoline
during the compliance period, which had the following oxygen contents:

Batch       Gallons           % Oxygen               Oxygen Content Units

2           100,000           2.0%                    200,000

3           100,000           2.3%                    230,000

4           100,000           2.9%                    290,000

      In this example, the CAR had no bulk transfers of gasoline to another control
area, or to any non-control areas.  Also, it is assumed that all the gasoline associated
with these four batches was sold or dispensed in this same control area during the
same control period. Therefore, the four batches of gasoline received constituted the
total gasoline which was relevant to the oxygenated fuel compliance determination.  To
determine compliance, the CAR compares the required total oxygen credit units to the
actual oxygen credit units which resulted from the gasoline sold or dispensed into the
control area.

      The  required oxygen credit units are calculated by multiplying the averaging *
standard times the total volume in gallons.  The averaging standard is 2.7 weight
percent oxygen, meaning that in this example, the resulting  required total content of
oxygen  is:

      2.7 X 400,000 gallons  = 1,080,000 oxygen credit units

      The  actual oxygen credit unit total is compared to this required total.  In this
example the actual total is 1,020,000 oxygen credit units, which is 60,000 oxygen
credit units less than the required total.  As a result, the CAR must Obtain  60,000     -:
oxygen  credit units generated by another CAR in the same control area and averaging
period in order to achieve compliance.

      For each control area  served by a CAR, calculations  such as these must be
computed.
                                       17

-------
      The next sample calculation demonstrates how a CAR or terminal operator will
compute the running weighted average oxygen content of a single bulk tank out of
which oxygenated fuel is sold or dispensed into any control area. These calculations
shall be used by a QAR or terminal operator who receives oxygenated gasolines  of
varying oxygen content during the compliance period, and  stores them all in the same
tank.  This example  is unlike the one above which would require that each shipment of
oxygenated fuel be handled separately.

      On day one of the control period, the CAR has 400,000 gallons of oxygenated
gasoline in a tank, which contains 2.0% oxygen by weight.  No gasoline is sold or
dispensed out of this tank on day one, and on day two, the CAR receives another
100,000 gallon shipment of oxygenated gasoline, this time containing 2.4% oxygen by
weight.  The running weighted average oxygen content of this tank, now containing
500,000 gallons of oxygenated gasoline, would be calculated as  follows:
      2.0 X 400,000 = 800,000 oxygen credit units
      2.4 X 100,000 = 240,000 oxygen credit units
The average running weighted average oxygen content is found by dividing the total
oxygen credit units in the tank by the number of gallons of oxygenated gasoline in the
tank:

      1040,000 divided by 500,000  =  2.08

Therefore, the running weighted average oxygen content of this tank is 2.08% oxygen
by weight..

      This calculation is a simplification.  It assumes that the amounts of gasoline
entering and leaving a tank can be easily accounted for, and that the different
oxygenate-containing gasolines will mix uniformly in the tanks.  In reality, some tanks
are capable of receiving and dispensing gasoline at the same time. Also, there may
be some stratification of the oxygenate levels within a tank.  But because the Agency
feels that there will be few instances  of terminal storage tanks receiving gasolines of
different oxygenate and oxygen contents, the RWOC system is the most appropriate
method.

      To continue the example, on day three the CAR dispenses 5 separate batches
of 10,000 gallons of oxygenated gasoline each from this tank into 5 separate trucks,
for a total of 50,000 gallons dispensed into the control area.  The gasoline in these
trucks has an oxygen  content of 2.08% by weight, based on the calculation above.
These withdrawals leave 450,000 gallons of oxygenated fuel in the tank.

      After dispensing this gasoline, the tank receives a shipment of 200,000 gallons
of oxygenated gasoline containing 2.7% oxygen by weight, bringing the total gallonage


                                       18

-------
in the tank up to 650,000 gallons.  The running weighted average oxygen content of
the tank after this addition would be calculated as follows:

      2.08 X 450,000 = 936,000 oxygen credit units
      2.70 X 200,000 = 540,000 oxygen credit units
      1,476,000 divided by 650,000 = 2.27

      Therefore, the running weighted average oxygen content of the tank after both
the dispensing of the 50,000 gallons and the addition of the 200,000 gallons of 2.7%
gasoline is 2.27% oxygen by weight.  Any gasoline subsequently dispensed into trucks
would have an oxygen content of 2.27% by weight.

      The next example is a compliance calculation which would be used by a
blende'r CAR. On day one of the compliance period the blender CAR received 900
gallons of gasoline containing 0.0% oxygenate by volume.  The blender CAR then
added 100 gallons of ethanol, bringing the total volume of gasoline to 1,000 gallons,
the oxygenate volume percentage up  to 10.0% and the oxygen content by weight up
to 3.5%.  The credit status of batch 1  of gasoline is calculated as follows:

Actual Oxygen Credit units = weight percent X gallons
              3.5 X 1,000 =  3,500 oxygen credit units

      The blender CAR had a total of three other shipments of oxygenated gasoline
during the compliance period, which had the following volumes and oxygen contents
after the blender added oxygenate to  the products:

                        % Oxygen
Batch       Gallons     when dispensed   Oxygen Credit Units

2           1,000          3.5%                 3,500

3           1,000          2.2%                 2,200

4           1,000          2.7%                 2,700

      In this example, the blender CAR had no transfers of gasoline to another control
area or to any non-control areas.  Also, it is assumed that all the gasoline associated
with these four batches was sold or dispensed in the same control area during the
same control period. Therefore, the four batches of gasoline received constituted the
total gasoline which was relevant to the oxygenated fuel compliance determination. To
determine compliance, the blender CAR compares the required oxygen credit unit total
to the actual oxygen credit unit total which resulted from the addition of ethanoi to the
gasoline which was sold or dispensed into the control area.
                                      19

-------
      The required oxygen credit unit total is calculated by multiplying the averaging
standard times the total volume in gallons. The averaging standard is 2.7 weight
percent oxygen, meaning that in this example, the resulting required oxygen credit unit
total is:

      2.7 X 4,000 gallons = 10,800 oxygen credit units

      The actual oxygen credit units total Is compared to this required total. In this
example, the actual total is 11,900 units, which is 1,100 units more than the required
total. As a result, the blender CAR may transfer 1,100 oxygen credit to another CAR
or blender CAR in the same control area and averaging period.

Attestation Engagements
    . .EPA's guidelines provide that each CAR which engages in credit averaging
shall, as part of the reporting requirements, engage a certified public accountant (CPA)
or firm of certified public accountants to perform an agreed-upon procedures
attestation engagement of the underlying documentation that forms the basis of the
reports, in accordance with the requirements of the guidelines, at the conclusion of
each annual control period. These guidelines permit the use of two different types of
auditors:                                         .

      1)     Internal auditors may be  used in certain situations.  Many large
companies may have internal auditors on staff who may perform the attestation
engagements required under these guidelines.  Internal  auditors may perform these
duties provided they are either Certified Internal Auditors (CIAs) or members in good
standing of the Institute of Internal Auditors, Inc. (I I A). As such, the internal auditors
will be bound by the IIA's Standards for the Professional Practice of Internal Auditing.
This requirement is intended to guarantee a number of things: That the internal
auditor is independent of the activities being audited under these guidelines; that the
internal  auditor is objective and proficient is his or her profession;  and that the internal
auditor is bound by a code of professional ethics.

      2)     When internal auditors meeting the criteria discussed above are  not
available within the CAR'S organization, an independent CPA or firm of independent
CPAs must be used to perform the attestation engagement required under these
guidelines.

      Regardless of whether an internal auditor or a CPA is used by a CAR,
attestation engagements would be conducted in accordance with  the applicable
professional standards following agreed upon procedure contained in these
guidelines. These guidelines cover a wider range of services than the audit of
historical financial statements.  EPA has worked closely with the AICPA and industry in
establishing these procedures which will accurately and efficiently provide the intended
compliance information in the most cost-effective manner. The IIA is supportive of

                                       20

-------
these procedures as well.

      These attestation engagements are not intended as substitutes for enforcement
audits conducted by the state, but are intended to serve as a means of improving
compliance with the oxygenated gasoline program by identifying problem areas to the
regulated parties. Such attestation engagements will assure regulated parties that
their records will be reviewed and cross-checked for accuracy by a third party.  They
will lead to the correction of simple arithmetic errors and will help to clarify the
regulatory requirements. And importantly, they will deter the submission of false
reports.                                       .

      EPA's guidelines provide that annual attestation engagements be conducted by
an internal auditor or a CPA with the report submitted by the CAR to the state no later
than 120 days following the end of the control period. Submission of the auditor's or
CPA's report by the CAR to the state is required, and failure to meet this requirement
would constitute a reporting violation by the CAR.  EPA intends to develop
standardized CAR-reporting forms for the state to adopt as a means of enhancing
uniformity and consistency in the attestation engagement and agreed-upon
procedures.  Drafts of these CAR-reporting forms have been sent to those states that
have expressed an interest  in establishing an averaging program and have been
placed in  Docket A-91-04.

      EPA believes that the objectivity of the attestation engagement, whether
performed by internal auditors or independent CPAs, is important for the effective
enforcement  of the oxygenated gasoline program. This objectivity will help to ensure
accurate accounting of oxygen content and gasoline volumes. In response to
comments, EPA has made several changes that should significantly reduce the costs
of these attestation engagements. These changes are: a) the use of specific "agreed-
upon" procedures; b) the use of internal audit staff, as appropriate and consistent with
both the IIA Standards  for the Professional Practice of Internal Auditing and the AICPA
Statement on Standards for Attestation Engagements; c) the use of audit sampling; and
d) requiring the report to be submitted in 120 days instead of 60 days after the  end  of
the control period.

      In order to maintain consistency within the process  EPA specifies the following
credentials for the CPAs to  be chosen by the regulated parties, and  provides the
following minimum guidelines to be followed in each attestation engagement.

                        (1) Credentials of internal auditors or CPAs.  The guidelines
                                       21

-------
require that the attestation engagements be conducted by either internal auditors or
independent CPAs and that attestation engagements are to be conducted in
accordance with either the Codification of Standards for the Professional Practice of
Internal Auditing or the Statement on Standards for Attestation Engagements,
respectively.  In general, the attestation standards deal with the need for technical
competence,  independence in mental attitude,  due professional care, adequate
planning and  supervision, sufficient evidence, and appropriate reporting.

      EPA's guidelines, in stating that the attestation engagements be performed in
conformity with the above-referenced standards, anticipate that the internal auditors
and CPAs will perform all of the required engagement procedures. EPA also expects
that the internal auditors and CPAs will document the procedures and findings with
working papers, consistent with the applicable  standards.

                        (2)   Attestation Guidelines in General. The guidelines
contain a listing of the general types of standard industry records which are required
to be included in the CPA's  attestation engagement procedures.

      The CAR shall, as part of the reporting requirements, conduct an attestation
engagement.  This engagement shall be performed by either an internal auditor or a
CPA or a firm of CPAs. This attestation engagement will study the underlying
documentation that forms the basis of the reports, in accordance with the
requirements  of the guidelines, at the conclusion  of each annual control period.

      In performing the attestation engagement,  the CPA should determine the
sample size for each population to which agreed-upon procedures will be applied in
such a manner that a sample size is equal to that which would result by using the
following sampling parameters:

            Number in Population fN)               Sample Size
                  66 or larger.                         59
                  41  -  65                            41
       '          26  -  40                            31
                  0-25                      N or 24, whichever is
                                                      smaller

      The number of populations from which samples should be drawn will vary
depending on the circumstances.  Sample items  should be selected in such a way that
the sample can be expected to be representative of the population.

      If the internal auditor  or CPA agrees to use some other form of sample
selection and some other method to determine the sample size, that agreement
should be sum%arized in the internal auditor's  or CPA's report.
                                      22

-------
      In the case of an independent CPA, while an  CPA should be required to issue
the attestation engagement reports called for by the guidelines, nothing shall preclude
the CPA from requesting assistance from the CAR'S internal auditors in accordance
with the Statement on Standards for Attestation Engagements.

                         (3)   Attestation Engagement Guidelines for Control Area
Responsible Parties. It is EPA's belief that many CARs will also be terminal operators.
However, not all CARs will be terminal operators and therefore all CARs may not have
access to some of the records referenced below.  For example, a non-terminal
operator CAR will likely not possess records showing the oxygen content of gasoline
entering the terminal. The requirements applicable to non-terminal operator CARs and
blender CARs would therefore be less exhaustive than those listed below.  These
parties would have to demonstrate the basis of their compliance calculation.

      Following is a listing of the records each CAR should maintain:

      1.    Records which show the quantity and oxygen content of oxygenated
            gasoline entering the terminal and leaving the terminal in bulk;
      2.    Records which show the destination, quantity and oxygen content of
            truckloads of oxygenated gasoline going to specific control areas;
      3.    Records which show the oxygen content of gasoline in storage tanks
            from which trucks are loaded, and the calculations which formed the
            basis for claimed oxygen content;
      4.    Testing results for storage tanks when additional gasoline is added;
    .  5.    Records showing the oxygenate type and amount which was blended;
            and
      6.    Records which show the beginning and ending inventories and
            oxygen contents of all gasoline and oxygenate storage tanks involved
            in the oxygenated gasoline program.

      Terminal operators normally prepare daily operations summaries for the
volumes  of each tank's inventory balances  (beginning and ending), transfers in and
transfers out.  Daily  reports are supported by pipeline meter tickets, truck tickets, and
tank gauging reports.  These daily reports are then summarized by month or quarter.

      The chemical characteristics of the product stored or moved into or out of each
tank are  based on periodic laboratory analysis or certificates of analysis from the
supplier. In order to comply with the guidelines, laboratory reports (or summaries
thereof) currently in  use must be revised to document more fully the oxygen content of
the oxygenated gasoline, and to provide a method of averaging these characteristics.
Compliance with the minimum 2.0% oxygen by weight requirement must be strictly
monitored.  The exact form of the detailed or summary reports has not yet been
determined, but the  prudent terminal operator will  likely perform computer analysis and
summarization of the data. These reports will also be the basis for calculating

                                      23

-------
compliance with the oxygen standard, and determining the amount of credits
generated or required.

      Special circumstances for terminals will likely require special data to be
collected in order for the CAR to demonstrate compliance, credit generation, or debit
generation.  Each CAR is responsible for assuring that such data is available.

                         (4)   CAR Reporting Forms.  The guidelines provide that
the internal auditor's  or CPA's agreed-upon procedures be applied to standardized
forms provided by the state.

Refiners and  Importers
      Refiners and importers are responsible for determining the oxygen content of all
gasoline produced or imported.  This determination must be made separately for each
batch of gasoline. The importance of correctly determining the oxygenate content of
each batch of gasoline is that this  parameter must be known when the gasoline arrives
at the control area where it will be dispensed. The shipping documents which
accompany each batch of  gasoline down the distribution chain must specify the
oxygenate and oxygen contents associated with each  shipment of gasoline.  In this
manner, the person who brings the gasoline into the control area where it is intended
to be used knows the oxygenate and oxygen contents for which an accounting must
be made.

      An effective state oxygenated  gasoline program would include state inspections
and audits of gasoline refiners and importers. The purpose of these inspections and
audits is to collect and analyze samples of gasoline stored at the refinery or import
facility to determine if the gasoline has been properly tested and classified. In
addition, the states shall audit testing records for oxygenated gasoline previously
produced or imported for proper classification and oxygen content.

      To facilitate these audits EPA's guidelines provide that refiners and importers  be
required to retain copies of documents which demonstrate that appropriate sampling
and testing was conducted to support the oxygen contents claimed. The guidelines
also provide that refiners and importers  retain copies of documents which describe the
purchase or production of  oxygenated gasoline as additional support for oxygen
content.

      These  records are to be retained at the refinery or import facility if practicable,
or at the business office of the refiner or importer. EPA recommends that states
establish a record retention requirement which coincides with their relevant statutes  of
limitations for enforcement of their oxygenated gasoline programs.

      Where a violation is found at a refinery or an import facility the refiner or
importer would be solely liable. The refiner or importer would have no specified

                                       24

-------
defense where the violation is discovered at that facility other than to contest the
existence of the violation.

      EPA's guidelines provide that in cases where gasoline produced or imported by
a refiner  or importer is found downstream from that party for which the oxygen content
of the gasoline is improperly stated, the refiner or importer will be presumptively liable
for these violations. The rationale for this presumption is discussed above.  The
refiner or importer will be able to avoid liability if it could demonstrate that it did not
cause the violation and that the results of tests performed by the refiner, importer or
blender on the gasoline show that the proper classification and oxygen content of the
gasoline  was recorded when it left the control of the refiner or  importer.

      In  cases where gasoline which is identified by the corporate, trade or brand
name of  a gasoline refiner is improperly classified or for which the oxygen content is
improperly stated, the guidelines provide that the named refiner be presumptively-
liable. This liability would attach regardless of who actually produced or imported  the
gasoline  (e.g., the named refiner would be presumptively liable even though the
gasoline  was obtained by the named refiner from another refiner through an exchange
agreement). In order to avoid liability in this situation, these guidelines specify that the
named refiner must show the following:

      (1)   Records of test results for the gasoline when it was produced or  .
            imported showing the oxygen content; and
      (2)   The violation was caused by action (s) of someone other than  the refiner
            or its  employees or agents; and
      (3)   The violation was caused by an act in violation of law, or an act of
            sabotage or vandalism; or
      (4)   The violation was caused by an act which was in violation of a
            contractual obligation designed to prevent such violations which was
            imposed by the refiner on the party operating under the refiner's brand
            name, and despite periodic sampling and testing by the refiner to assure
            compliance with the contractual obligations; or
      (5)   The violation was caused by the act of a carrier or other distributor
            engaged by the refiner for transportation of gasoline but with whom
            the refiner did not have a contractual relationship, despite efforts by
            the refiner (such as a periodic sampling and testing) designed to
            assure that violations do not occur.
                                  '
      This refiner's defense for violations found at branded facilities is closely
modeled upon the enforcement schemes followed in the federal gasoline lead
contamination, volatility, and diesel fuel sulfur content regulations.
                                       25

-------
Distributors
      EPA believes that gasoline distributors should be responsible for ensuring that
gasoline sold, dispensed, transported or stored by a distributor downstream of the
terminal is  properly characterized as either oxygenated gasoline or nonoxygenated
gasoline.  Distributors would be prohibited from selling, storing or transporting
gasoline intended for use in a control area during the control period which does not
meet the 2.0% minimum oxygen content requirement.  Distributors would not be .
prohibited from storing nonoxygenated gasoline within the control area as long as it is
intended for sale in a non-control area or is intended for sale outside  of the control
period or is to be used for blending with ethanol and is properly segregated and
labeled.  If the fuel is intended for use in the control area and is to be sold or
dispensed  after the end of the control period in the control area then  the storage tank
should remain segregated until that time.

      EPA's guidelines provide that a distributor downstream of the terminal should
be liable for violations of the above  requirements found at the distributor's facility.  In
addition, the guidelines specify that distributors should be liable for such violations
found at facilities downstream from  the distributor, which could include facilities
operated by other distributors, downstream carriers, retailers and/or wholesale
purchaser-consumers.

      In the case of oxygenated gasoline which is sold, transported,  or stored
between the refinery or import facility and a control area terminal, these guidelines
provide that distributors have the additional responsibility of ensuring  that their
gasoline conforms to the oxygen content which is stated  in the paperwork which
accompanies the gasoline.  In EPA's guidelines,  distributors would  be liable for
violations of this requirement found  at the distributor's facility, and for violations found
between the distributor and the control area terminal or oxygenate  blending facility.

      Under EPA's guidelines, the distributor upstream of a control area terminal or
oxygenate  blending facility could avoid liability for the above requirements if it could
show:  (1)  that it or its employees or agents did not cause the  violation (e.g., by
showing causation elsewhere); (2) possession of documents required to accompany
the gasoline, such as invoices or bills of lading, which contain the information required
by paragraph (h) of the guidelines;  and (3)  evidence of a quality  assurance sampling
and testing program carried out by  the distributor to monitor, when appropriate, the
oxygen content.

      EPA believes that when gasoline found at a distributor's facility is improperly
classified or the  oxygen content is not properly stated in the accompanying
paperwork, persons  upstream from the distributor should be presumptively liable for
these violations. The upstream parties would  include refiners, importers, blenders,
carriers and distributors, with the exception that liability associated  with the minimum
oxygen content  requirement would  not apply upstream of the control area terminal.

                                       26

-------
Carriers
       Carriers are distinguished from other distributors in that carriers do not take title
to the product they store or transport.  As a result of this distinction, carriers
traditionally, have had liability presumptions and defenses which are different from
other distributors under federal fuels enforcement fchemes (e.g., volatility, leaded
contamination, and diesel sulfur).

       There are at least two options for ensuring that oxygenated gasoline
transported or stored by upstream carriers and downstream carriers conforms to the
oxygenated gasoline requirements.  One option is to make carriers liable only for
violations detected at the carrier's facility, unless the carrier is able to show  that it did
not cause the violation. Under this option, carriers would not be presumptively liable
for violations found downstream from the carrier's facility, unless it can be shown that
the carrier in fact caused the violation.  This is the traditional approach used for
carriers.

       The second option is to make carriers presumptively liable for violations
detected downstream from the carrier. The Agency feels that it is possible for carriers
to alter the quantity or quality of the product they handle, thereby necessitating some
form of quality assurance.  A carrier could cause a violation under these guidelines, for
example,  if its holding tanks were not properly emptied and cleaned between
shipments, or if it inadvertently mixed oxygenated gasolines with nonoxygenated
gasolines. Carriers would be able to avoid liability by showing that they did not cause
the violation, and by showing that the gasoline they transport or store conforms to the
accompanying shipping documents.

       The rationale for the first liability scheme is that carriers normally do not alter the
quality of the gasoline they transport or store - in fact, the EPA's definition of carrier in
40 CFR Part 80 requires that they not alter the quality of the gasoline.  Under this
argument, carriers only transport or store what they are given, and have no control
over the product. This approach was found to be most appropriate in the gasoline
volatility program because EPA is able to sample and test gasoline at any point
downstream from the carrier to determine if the gasoline conforms to the standard.
When violations of the applicable volatility standard are found, EPA normally is able to
gather enough facts to establish who caused the violation, thus deterring future
violations.

       Because carriers by definition do not alter the quality or quantity of the gasoline,
and because an extensive testing requirement would be an undue burden on the
industry, EPA believes that applying presumptive liability to carriers is not appropriate.
Retailers and Wholesale Purchaser-Consumers
                                        27

-------
      The guidelines provide that during the control period, retailers and wholesale
purchaser-consumers in CO nonattainment areas are prohibited from selling or
dispensing gasoline that has less than the required 2.0% minimum oxygen for use in a
control area.  Retailers and wholesale purchaser-consumers would be liable for
violations of the above requirements found at their facilities.

      Under various federal fuels enforcement schemes, retailers and wholesale
purchaser-consumers have been able to avoid liability by showing they did not cause
the violation.  The same shall hold true for these guidelines, so long as the facility also
has retained all documents which accompanied the gasoline in question.

Product Transfer Documentation
      Under these guidelines, on each occasion physical custody or title of gasoline
transfers from one* party to another (other than when gasoline is sold or dispensed for
use by the ultimate consumer at a retail outlet or wholesale purchaser-consumer
facility),  certain information shall be exchanged to facilitate  administration and
enforcement. This information should be recorded either on a separate transfer
document or through the addition of the required information to paperwork which
already accompanies the shipment of gasoline.  This information should include the
following:
                  a.     The date of the transfer;
                  b.     The name and address of the transferor;
                  c.     The name and address of the transferee;
                  d.     The volume of gasoline which is being transferred;
                  e.     The proper identification of the gasoline as
                         nonoxygenated or oxygenated;
                  f.      The location of the gasoline at the time of the transfer;
                  g.     The type of oxygenate; and
                  h.     The oxygen content of the gasoline, and for transfers
                         upstream of the control area terminal and for transfers
                         between CARs, also the oxygenate volume of the
                         gasoline.

Recordkeeping and Reporting
    All CARs should be required to maintain reports  containing compliance information.
Parties which have selected the  option of meeting the standard on a "per-gallon" basis
would be required to maintain a basic set of information. The records kept by these
parties would be much simpler than those the guidelines require of averaging parties. •
Information to be recorded would include data on product  received by the party  (for
example, the date the product was received, the source of the shipment, the type of
product received, the total volume of the shipment),  and data on the product sold or
supplied by the party (for example, the date the product was sold or supplied, the type
of product soldsor supplied, the  total volume of the shipment, the name of the  person
to whom the product was sold or supplied, the oxygenate  content, and the oxygen


                                       28

-------
content of the product).

      Parties which have selected the option of meeting the standard on an average
basis would be required to maintain more detailed information because of the greater
complexities of demonstrating compliance when averaging and credit trading are
involved. Information to be recorded would include data on product received by the
party (for example, the date the product was received, the source of the shipment, the
type of product received, the total volume of the shipment, and the results of any tests
performed), data on the product sold or supplied  by the party (for example, the date
the product was sold or supplied, the type of product sold or supplied, the total
volume  of the shipment, the name of the person to whom the product was sold or
supplied, the oxygenate content, and the oxygen  content of the product).  In addition,
the party would also be  required to calculate the average oxygen content of its
product based on such information and according to the procedure outlined above.

      In addition to the  information detailed above,  CARs engaging in trading  oxygen
credits during a control period would be required  to supply additional information in
their reports. Such information should include the name, CAR identification number
and address of the other party in each trade and  the quantity of oxygen credits
(volume and oxygen content of gasoline) traded.  The party selling or otherwise
transferring oxygen credits shall demonstrate how the credits in question were
calculated.  The party buying or otherwise receiving oxygen credits would be required
to calculate its compliance with the regulatory standard through the use of these
credits.  Both parties to an oxygen credit trade would submit documentation to the
state which is adequate to demonstrate their mutual agreement to the trade and to
transfer the credits no later than 30 working days after the close of the averaging
period for which the trade is reported. A contract signed by both parties no later than
30 days after the close of the relevant averaging period would be sufficient for this
purpose. A purported trade would  not be recognized as valid unless both parties
have reported and adequately documented it.

    Persons who own control area terminals but who do not own the gasoline which is
dispensed from those terminals are not subject to the averaging standard.  These
terminal operators would be required to maintain  records.  These records would have
to include information on the ownership, volume,  and oxygen content of gasoline sold,
dispensed or transported during each averaging period, and the location to which it
was transported and whether it was within a control area or not.  Such records would
provide a partial cross-check on reports submitted by persons subject to the
regulatory standard.

    All parties subject to these recordkeeping requirements should be required to
retain these records for the period of time established by the state. The records
should be available for appropriate  state review, even if the party is not required to
submit information to the state.  For all records, the state would have the authority to


                                       29

-------
determine whether any record should be recognized as meeting regulatory
requirements or not.

    The only parties which would be required to send in compliance reports to the
state are the CARs.  No later than 30 days after the close of the averaging period,
each CAR should submit a report to the state evidencing compliance and detailing its
purchases, shipments, sales, and credit accounting for the averaging period in
question.  Also, no later than 120 days  after the close  of the averaging period, each
CAR should submit an internal auditor's or CPA's attestation engagement report in
accordance with these guidelines.

Sampling and Testing Methodologies
      The sampling methodologies recommended for oxygenated fuels programs are,
in part, the same as those set forth at 40 CFR Part 80, Appendix D relating to
sampling procedures for fuel volatility.

      In these guidelines, the Agency is identifying two separate testing methods for
eventual use with these oxygenated gasoline guidelines. The American Society for
Testing  and Materials (ASTM) standard  test method, Designation D 4815-89, is
presently the most widely-used method  for the determination of alcohols and MTBE in
gasoline by gas chromatography. This  test method covers a procedure for
determination of methanol,  ethanol, isopropanol, n-propanol, isobutanol, sec-butanol,
tert-butanol, n-butanol, and methyl tertiary butyl ether (MTBE) in gasoline by gas
chromatography.  However, it does not  currently have the capability to detect the
presence of some of the heavier oxygenates  in gasoline, one example being TAME,
although ASTM is planning to extend the scope to include up to 15% MTBE by volume
and 17% TAME by volume.  Adaptation of the method 4815 for ETBE analysis is
straightforward and merely requires a change of internal standards. An additional
shortcoming of ASTM D4815-89 is that  many states which have already implemented
oxygenated gasoline programs have found the ASTM precision standards to be
inadequate because of large variations in accepted oxygen level measurements.

      ASTM is currently focusing on both of these issues,  and although a revised test
method  is not yet available for industry  use, the ASTM committee has allowed EPA to
publish a working final draft of the revised test method, ASTM D-4815-9X. This
method  has an expanded range of oxygenates which it can test and it will yield results
in terms of mass percent oxygen. This  draft  is included in these guidelines  as
Appendix B. This draft has not yet been finalized by the ASTM committee, and there
are sure to be  some minor changes, but is included as part of this notice in order to
give advance EPA approval for the soon-to-be finalized method. Until this method is
finalized, ASTM D4815-89 is still in effect.12
      12  For ASTM  D-4815-89,  see 57  FR 4437  (February 5,  1992).

                                     30

-------
      Also in reaction to the shortcomings of ASTM D4815-89, EPA's laboratory in
Ann Arbor, Michigan has developed an alternative testing methodology.  This method
is a single column, direct injection gas chromatographic procedure for quantifying the
oxygenate content of gasoline.  Unlike the current ASTM method this method can be
used to detect all types of oxygenates in gasoline.  This method is included in these
guidelines as Appendix C.

      The Agency believes that the EPA test  (ORD) has various advantages over the
ASTM method, and is therefore preferable.  The Agency anticipates it to be more
accurate and easier to conduct. Like the new ASTM method, the OFID method will be
calibrated to yield its results in terms of the mass percent oxygen contained in the fuel
blend being analyzed, which is  simpler to convert into the unit of measurement of the
statutorily mandated standard than volume percent oxygenate is.  This fact will
eliminate the need for the detailed conversions described in the section below,
"Oxygen Content Conversions."

      However,  at this point in time the EPA OFID test is new, and industry is
understandably apprehensive about adopting  a relatively new test. These guidelines
therefore specify that either the ASTM test method or the OFID method may be used.
This will allow the regulated parties to use equipment they may already possess, and a
test method with which they are already familiar.  Nevertheless, the Agency would like
to strongly recommend use of the Ann Arbor  method.

      In addition to the approval of these two testing methods, EPA would like to
establish a procedure whereby -additional testing methods may be approved by the
Agency.  EPA recognizes that there are many potential tests for use in the detection  of
oxygenates to gasoline, and would like to encourage the development of even newer
and more efficient methods. The Ann Arbor lab has already begun evaluating some
alternative laboratory methods,  as well as some oxygenate screening devices which
may be useful in the field.  Therefore, the Agency shall work on creating a procedure
for the evaluation and approval of other oxygenate tests.

      The ASTM method contains precision information for the volume percent of
various oxygenates that varies as a function of the volume of oxygenate being
measured.  The Agency plans to use a single testing tolerance for ethers and  alcohols
that represents the predominant volume of these oxygenates that is expected  to be
used to comply with the oxygen content requirements. The use of a single testing
tolerance for each oxygenate will simplify enforcement. This tolerance, as mandated
by the Act, section 211 (m) (2) (B), will be established by the Administrator.

Oxygen Content Conversions
      An issue has been raised concerning the ability to accurately determine the
oxygen content of gasoline when oxygenates are added by volume (usually
downstream of the refinery). This is a concern because as the specific gravity (or


                                      31

-------
density) of the base gasoline varies, the weight fraction of oxygenate (and oxygen)
varies for any specific oxygenated gasoline blended.  Hence, two blends of oxygenate
could result in differing oxygen weight fractions if the specific gravity of the base
gasolines for the two blends differ.

      Typically, oxygenates are blended with gasoline volumetrically.  For example, a
"ten percent ethanol blend" typically refers to a volume percent.  The standards of an
oxygenate program as delineated in the Clean Air Act Amendments are in terms of
weight percent oxygen.  Technically, in order to calculate the weight percent oxygen in
the oxygenate blend, several factors must be taken into consideration. These are:
temperature and specific gravity of the oxygenate and the gasoline, and, for ethanol,
the amount of denaturant,  which is  some fraction of the volume ethanol added to the
gasoline.  Elsewhere in these guidelines, it is stated that standard temperature will be
60 degrees Fahrenheit.  In order to calculate the weight percent oxygen in the blend,
the weight percent oxygenate must be calculated.  Accordingly, to calculate the weight
percent oxygenate from  volume percent oxygenate, specific gravities of the oxygenate
and the blend  must be taken into consideration.  (Specific gravities (or densities) as
well as'weight percent oxygen in the oxygenate may  be found in Table 1 for common
fuel oxygenates.)

Table 1.  Specific Gravity and Weight Percent Oxygen of Common Oxygenates
Oxygenate
methyl alcohol
ethyl alcohol
normal propyl alcohol
isopropyl alcohol
normal butyl alcohol
isobutyl alcohol
secondary butyl alcohol
tertiary butyl alcohol
methyl tertiary butyl ether (MTBE)
tertiary amyl methyl ether (TAME)
diiso propyl ether (DIPE)
ethyl tertiary butyl ether (ETBE)
Weight fraction
oxygen
0.4993
0.3473
0.2662
0.2662
0.2158
0.2158
0.2158
0.2158
0.1815
0.1566
0.1566
0.1566
Specific Gravity
at 60 deg F
0.7963
0.7939
0.8080
0.7899
0.8137
0.8058
0.8114
0.7922
0.7460
0.7752
0.7300
0.7452
                                       32

-------
The following equation describes the conversion from volume percent oxygenate to
weight percent oxygenate:
                                   =V        x  ^senate                     M\
                           oxygenate v oxygenate*    j                          »*'
Where

      -W          = weight fraction (for percent, multiply by 100)
      oxygenate   = oxygenate in the blend
      bl           = blend
      V           = volume fraction
      d           = specific gravity.

The specific gravity of the oxygenate is known (see Table 1) and, if the specific gravity
of the blend has been measured and is, therefore, known, the calculation is
straightforward.  If, however, the specific gravity of the blend is unknown, it can be
estimated as the volume weighted contribution of .the specific gravities of the gasoline
to which the oxygenate is added and the oxygenate itself:
Where

      gas = gasoline to which oxygenate is added.

The weight fraction of oxygen in the blend is simply the product of the weight fraction
of oxygen in the oxygenate (from Table 1) and the weight fraction of oxygenate in the
blend. Therefore, the weight fraction of oxygen in the blend is:
                         W     ~W       ^fW                                 f 3 ^
                         "oxygen "oxygenate  oxygen/ oxygenate                  ,  * '
Where

      oxygen/oxygenate = oxygen in the oxygenate.

Substituting equations (1) and (2) in equation (3), results in:
                                       33

-------
                    jf     _ *oxyyenat8X"oxygenate*"oxygen/'oxygenate
                            * gas^^gas'  * ^oxygenate*® oxygenate'



 For blends with more than one oxygenate, the equation becomes:
                          TV T7        y/-f       yltf             S
                  nr     _ *•** v oxygenate'^^oxygenate*'" oxygen/oxygenate'
                  "oxygen    / y  xfj  \ + V/V    :   Trf       \
                            v 'a^^' ^v *wiate*"oxygenate'
 The following examples demonstrate use of the equation:

 Question 1:  Suppose nine gallons of neat ethanol are blended with 91 gallons of
 gasoline to make 100 gallons of ethanol blend gasoline.  The specific gravity of the
. gasoline is 0.74.  What is be the weight percent oxygen in this blend?

 Answer 1: In this case, the volume fraction of ethanol is 0.09 and the volume fraction
 of gasoline is 0.91.  The specific gravity of neat ethanol (from Table 1) is 0.7939 and
 the specific gravity of the gasoline is stated to be 0.74.  Hence, the weight fraction of
 oxygen can be calculated using equation (4) as follows:


                       v        0.09x0.7939x0.3473                      (6)
                        oxy~ (0.91x0.74)+(0. 09x0.7939)
                                  ^=0.0333                              (7)
 Therefore the weight fraction of oxygen in such a blend is 0.0333 or 3.33 percent.

 Question 2:  Suppose 1000 gallons of MTBE are blended with 6000 gallons of gasoline
 to make 7000 gallons of MTBE blend gasoline. The specific gravity of the gasoline is
 0.75. What is be the weight percent oxygen in this blend?

 Answer 2:  In this case, the volume fraction of MTBE is 1000/7000  or 0.1429 and the
 volume fraction of gasoline is 6000/7000 or 0.8571. The specific gravity of neat MTBE
 (from Table 1Vis 0.7460 and the specific gravity of the gasoline is stated to be 0.75.
 Hence, the wfl|ht fraction of oxygen can be calculated using equation (4) as follows:
                                       34

-------
                   LI         Q-.'l429xO.7460x0.1815                     (8)
                     oxy~ (0.8571x0.75) + (0.1429x0.7460)
                                 ^=0.0258                             (9)
Therefore the weight fraction of oxygen in such a blend is 0.0258 or 2.58 percent.
In the following example, multiple oxygenates are used.

Question 3:  Suppose 800 gallons of MTBE and 200 gallons of TAME are blended with
6000 gallons of gasoline to make 7000 gallons of blend gasoline. The specific gravity
of the gasoline is 0.73. What is be the weight percent oxygen in this blend?
 **

Answers: In this case, the volume fraction of MTBE is 800/7000 or 0.1143, the
volume fraction of TAME is 200/7000 or 0.0286 and the volume fraction of gasoline is
6000/7000 or 0.8571.  The specific gravity of neat MTBE (from Table 1) is 0.7460,  of
neat TAME is 0.7752 and the specific gravity of the gasoline is stated to be 0.73.
Hence, the weight fraction of oxygen can be calculated using equation (5) as follows:


                                                                         (10)
                                      0.0253                            (11)
Therefore the weight fraction of oxygen in such a blend is 0.0253 or 2.53 percent.

      While refinery blending of oxygenates presents little problem in calculating the
oxygen weight percent since the specific gravity of the gasoline blendstock is typically
measured on a routine basis, with terminal blending the specific gravity parameter may
not be readily available to an oxygenate blender. Hence, the Agency believes it may
be appropriate to provide a second option by which the oxygen content of oxygenated
gasoline blended at the terminal may be determined.  The following two variables must
be considered:  1)  What should be used for the specific gravity of the gasoline
blended with oxygenate at the terminal (which is variable), and 2) For ethanol blends,
what considerations should be made for the presence of a denaturant in the ethanol?
Gasoline samples from the 1990 Motor Vehicle Manufacturers Association (MVMA)
fuels database indicate an  average specific gravity of 0.742029 with a standard
deviation of 0.013285. Using two times the standard deviation to create a lower and


                                      35

-------
upper bound (assuming the vast majority of samples lie within this range), a range of
oxygen weight percents can be calculated for an upper end, lower end, and average
gasoline specific gravity using equation (4). Table 2 shows the results of such an
analysis and includes an analysis if one assumes the volume fraction of ethanol and
the weight fraction of ethanol to be equal.

Table 2:  Oxygen Weight Percents Based Upon Gasoline Specific gravity
            Ethanol: 10 volume %/No denaturant
Description
W % eth = V % eth
High End Specific gravity
Average Specific gravity
Low End Specific gravity
Gasoline Specific gravity
0.7939
0.7686
0.7420
0.7155
Weight % Oxygen
3.47
3.58
3.69
3.81
      If the assumption is made that 5 percent by volume of the ethanol is denaturant
(i.e., 0.5 percent by volume of the final blend is denaturant) and therefore the ethanol
volume contribution to the final blend is 9.5 percent, the following results apply:

Table 3:     Oxygen Weight Percents Based Upon Gasoline Specific gravity
            Ethanol: 9.5 volume %/Denaturant: 0.5 volume %
Description
W% eth = V % eth
High End Specific gravity
Average Specific gravity
Low End Specific gravity
Gasoline Specific gravity
0.7939
0.7686
0.7420
0.7155
Weight % Oxygen
3.30
3.40
3.51
3.62
      Although the Agency believes that little blending of oxygenates other than
ethanol is performed at the terminal, a similar analysis could apply for MTBE and/or
other oxygenates.  However, for oxygenates other than ethanol, the denaturant
consideration is not applicable. Table 4 shows such an approach for a 15% MTBE
blend.
                                      36

-------
Table 4:     Oxygen Weight Percents Based Upon Gasoline Specific gravity
            MTBE: 15 volume percent
Description
W% MTBE = V% MTBE
High End Specific gravity
Average Specific gravity
Low End Specific gravity
Gasoline Specific gravity
0.7460
0.7686
0.7420
0.7155
Weight % Oxygen
2.72
2.65
2.73
2.81
      Since the Agency believes that oxygenates blended at the terminal are blended
volumetrically and that most gasolines should be near the average specific gravity
listed above and most ethanol blends do contain 0.5 percent by volume denaturant,
Table 3 is most appropriate for 10 percent ethanol blends. Therefore, utilizing the
'average gasoline" row from Table 3, the appropriate level of oxygen associated with a
10 percent (by volume) ethanol blend is best estimated to be 3.51 weight percent.
Thus, the Agency proposes that one alternative for determining the oxygen content for
terminal-blended ethanol-gasoline blends is to simply assume a 3.51 weight percent
oxygen based on the above analysis.  Likewise, for a terminally blended  15 percent
(by volume) MTBE blend, the appropriate oxygen content would  be 2.73 weight
percent. For other volumes of these or other oxygenates, a terminal blender may
simply substitute the appropriate  values above for average gasoline specific gravity
and the values in Table 1 in equation 4 to calculate the appropriate oxygenate level.
As mentioned previously, lor refinery blended oxygenates, the actual measured
specific gravities should be utilized.

      The second alternative available to the terminal blender would be to actually
measuring the appropriate specific gravities. This option would benefit any blender
who feels that he is using a base gasoline which will  result in a higher oxygen content
by weight when tested than would be found through  use of the above calculations.
Although there has been some comment on the possibility that some blenders may
manipulate this  choice in order to always obtain the most favorable oxygen levels, the
Agency believes that the time and money associated with the performance of these
tests should discourage any party from testing every batch of gasoline, then using the
more favorable  number in its records.  The Agency feels that most parties will opt to
rely on the calculations presented here.

Purity and Oxygen Content
      Because many parties in the gasoline distribution network will be relying on  the
written records they receive from other parties in the network in order to determine the
amount of oxygenate contained in the fuel they offer  for sale, sell, store,  or dispense,
the issue of purity is an important one.  Fuels must not be represented as containing
more oxygenate than they actually do. The calculations provided above assume that
                                       37

-------
on the average, 0.5% of any 10% ethanol by volume blend is denaturant, making the
ethanol volume contribution to the final blend 9.5%.  Therefore, based on the Table 3
calculations, for terminal-blended ethanol gasolines one option is to simply assume a
3.51% oxygen by weight. The terminal operator or Blender CAR would always have
the second option available, which is to actually measure the appropriate specific
gravities. In either case, the purity should be noted on any records and transfer
documents created by the CAR.

Blending Allowance
       In order to allow for the dilution of oxygenates during transport and storage and
for the varying nature of the density of gasoline, the Agency is recommending the use
of a blending allowance for the measurement of all oxygenates which fall under the
"Substantially Similar" ("sub sim") Interpretive Rule.13 The allowance will permit the
blending of gasoline at  levels 0.2% percent oxygen by weight higher than allowable
under the "sub sim" interpretive rule. This allowance is desirable from a practical
standpoint since the requirement for program areas and the legal maximum under
"sub sim" are the same (2.7% oxygen by weight).  It will compensate for the dilution of
some oxygenates during transport and storage, providing some flexibility to gasoline
producers who are likely to blend gasoline at points upstream from terminals and
transport it to the terminal and to gasoline blenders at the terminal level.

       It is important to note that this allowance applies  only to oxygenates blended
under "sub sim." The allowance would not apply to oxygenates waived to oxygen
levels above 2.7 weight percent oxygen.  (Hence,  an ethanol blend could not be
blended to levels higher than that allowed under the "gasohol waiver.")

       In order to compensate for the problems associated with dilution and density,
EPA will exercise discretion in enforcing the maximum "sub sim" limit by permitting a
blending allowance of +0.2 percent oxygen by weight for all "sub sim" gasolines.  For
example, MTBE or TAME blends containing up to 2.9% oxygen by weight will be
considered acceptable when detected at any point in the gasoline distribution network.
This will allow producers and upstream blenders to blend slightly higher volumes of
oxygenate into their gasoline, thereby anticipating and avoiding the potential  loss of
oxygen in the gasoline intended for sale in an oxygenated gasoline program.  A similar
blending allowance was announced by EPA in  its  Federal Implementation Plan for the
Maricopa and Pima carbon monoxide nonattainment areas.

       It is important to note that this blending allowance is an enforcement discretion
only.  CARs may neither factor "sub sim" gasolines containing more than 2.7% oxygen
      13  56 FR 5352 (February 11, 1991).

      56 FR 5458 (February 11, 1991).

                                      38

-------
by weight into control period averages, nor generate credits based on "sub sirrT blend
gasolines containing more than 2.7% oxygen by weight.

      This blending allowance will be considered separately from the testing tolerance
which is to be established at a later date by the Administrator.

Approved Oxygenates
   An oxygenate is any substance which, when added to gasoline, increases the
amount of oxygen in that gasoline blend.  It is unlawful to introduce oxygenated
gasoline  into commerce unless it is either "substantially similar" to certification fuel in
accordance with § 2ll(f)(1) of the Act, or permitted under a waiver granted by the
Administrator under the authority of § 2ll(f)(4) of the Act. The following oxygenates are
currently approved. Others may be approved by the Agency in the future, at which
time they may be automatically recognized as approved under  these guidelines.

   Through a series of waivers and interpretive rules,  the Agency  has determined the
allowable limits for oxygenates in unleaded gasoline. The "Substantially Similar"
Interpretive Rule allows blends of aliphatic alcohols other than methanol and aliphatic
ethers, provided the oxygen content does not exceed  2.7% by  weight. An oxygenated
blend may contain any mix of alcohols or ethers (other than methanol) at levels up to
2.7% by weight.  It also provides for blends of methanol up to 0.3  percent by volume
exclusive of other oxygenates, and up to 2.75% by volume methanol with an equal
volume of butanol or alcohols of a higher molecular weight.

   The following individual waivers pertaining to the use of oxygenates in unleaded
gasoline  have  been issued by the Agency under the authority of § 211(f)(4), and are
available for use by all parties.

            1.  "  Blends of up to 10% by volume anhydrous ethanol (200 proof)
(commonly referred to as the "gasohol" waiver).15

            2.     Blends of methanol and gasoline-grade tertiary butyl alcohol
(GTBA) such that the total oxygen content does not exceed 3.5% by weight and the
ratio of methanol to GTBA is less than or equal to one. It is also specified that this
blended fuel must meet ASTM volatility specifications (commonly referred to as the
"ARCO" waiver).16

            3.     Blends of up to 5.0% by volume methanol  with a minimum  of 2.5%
by volume cosolvent alcohols having a carbon number of 4 or  less (i.e. ethanol,
    15 44 FR 20777 (April 6, 1979).

    16 44 FR 10530 (February 21, 1979).,
                                       39

-------
propanol, butanol, and/or GTBA). The total oxygen must not exceed 3.7% by weight,
and the blend must meet ASTM volatility specifications as well as phase separation
and alcohol purity and inhibitor specifications (commonly referred to as the "DuPont"
waiver).17

            4. Blends of up to 5.0% by volume methanol with a minimum of 2.5% by
volume cosolvent alcohols having a carbon number of 8 or less. The total oxygen
must not exceed 3.7% by weight, and the blend must meet ASTM volatility
specifications as well as phase separation and alcohol purity and inhibitor
specifications (commonly referred to as the "Octamix" waiver).18

            5.     Blends of up to 15.0% by volume methyl tertiary butyl ether
(MTBE), which must meet the ASTM D4814 specifications. Blenders must take
precautions that the blends are not used as base gasolines for other oxygenated
blends (commonly referred to as the "Sun" waiver).19

      It is the intent of these guidelines that oxygen content be calculated based upon
the actual content of oxygen  of a blend.   That is, the actual content of. oxygen in a
gasoline blend is determined based upon the volume of the oxygenate, excluding
denaturants or other non-oxygen-containing compounds.

Inability to Produce Conforming Gasoline Due to Extraordinary Circumstances
      Some parties suggested during the Regulatory Negotiation process that EPA
address the situation where extraordinary circumstances do not permit a regulated
party to comply with the requirements of a state oxygenated gasoline program under
Section 211 (m).  In appropriate extreme and unusual circumstances (e.g.,  natural
disaster or "Act of God") which are clearly outside the control of the refiner and which
could not have been avoided by the  exercise of prudence, diligence and due care,
states should consider allowing a refiner, for a brief period, to distribute fuel which -
does not meet the requirement for oxygenated gasoline if: 1) It is in the public interest
to do so  (e.g., distribution of the nonconforming fuel is  necessary to meet projected
shortfalls which cannot otherwise be compensated for); 2) The refiner exercised
prudent planning and was not able to avoid the violation and has taken all reasonable
steps to minimize the extent of the nonconformity; 3) The  refiner can show how the
requirements for oxygenated gasoline will be expeditiously achieved;  4) The refiner
agrees to make up the air quality detriment associated with the nonconforming
gasoline, where practicable; and 5) The refiner agrees to pay the state an  amount
    17 See 50 FR 2615 (January 17, 1985), 51 FR 15064 (April 22, 1986), 51 FR 39800
(October 31, 1986), and 52 FR 18736 (May 19, 1987).

    18 53 FR 3636 (February 8, 1988).

    19 53 FR 33846 (September 1, 1988).

                                      40

-------
equal to the economic benefit of the nonconformity minus the amount expended,
pursuant to number 4 above, in making up the air quality detriment.

IV.  Comment Summary and Response

      A.    Attestation Engagements
      Comments:  Several parties submitted comments on the issue of attestation
engagements to be performed by independent CPAs.  Some parties felt that outside
CPAs are not necessary and that the attestation requirements may be overly
burdensome, time consuming,  and expensive. Some parties suggested random
enforcement audits of selected facilities, if an audit requirement of this type is
necessary.  Some parties expressed support for attestation engagements by
independent CPAs as a justified enforcement tool.  It was  suggested that the cost of
attestation engagements can be reduced if EPA lets the independent CPA rely on
internal work consistent with the American Institute of Certified Public Accountants'
(AJCPA) guidelines.

      Response:   Due to cost considerations, EPA has revised these  guidelines to
allow some flexibility to some regulated parties.  EPA has  made several changes to
these guidelines in response to comments: a) These guidelines establish specific
"agreed-upon" procedures; b) the use of internal audit staff will be allowed as
appropriate; c) the guidelines establish a method of audit  sampling; and d) the
guidelines now require the attestation report to be submitted in 120 days instead of 60
days after the end of the control period.

      B.    State Jurisdiction^ Issues
      Comments:  Several parties expressed concern about the issue  of state
enforcement action against out-of-state parties.  These commenters tended to prefer
that EPA take the  enforcement "lead." Some commenters felt that states should be
permitted to delegate enforcement responsibilities for multi-state CMSAs and MSAs to
EPA.

      Response:   This issue  is not relevant to the scope of these guidelines, and is
therefore not addressed in these guidelines.

      C.    Inability to Produce Conforming Gasoline Due to Extraordinary
Circumstances
      Comments:  Most commenters support allowing marketers to distribute fuel
which does not meet the requirements for oxygenated gasoline in appropriate extreme
and unusual circumstances.  One commenter noted that relatively small shortfalls in
gasoline supply have historically lead to consumer panic and price increases.  Others
pointed out that some companies could be forced out of the market if there were no
provisions for relief.  One commenter pointed out  that any unfair economic advantage
could be mitigated if fines and  fees are applied. A few parties, responding to the July


                                      41

-------
9, 1991 notice, did not support the "variance" provisions and noted that, if variances
are available, they should be greatly restricted. Another commenter suggested that
the variance should be limited to 30 days.

      Response:  EPA agrees that "variances" are an appropriate response to
extraordinary circumstances. These "variances" should be severely restricted in
scope.

      D,    Recordkeeping
      Comments:  Some commenters expressed concern on the issue of how a CAR
can "assure" that oxygenated gasoline is delivered to the intended destination.  Also,
several commenters expressed concern about the extent of the recordkeeping
requirements.  Most commenters recommended a record retention period of 2 or 3
years.

      Response:  EPA has attempted to clarify the CAR'S responsibility by stating
that they can rely on the purchaser's representation provided that they put the
required information on the bill of lading. Moreover, a recordkeeping requirement
equivalent to each state's statute of limitations will allow the state to effectively pursue
violators. This is especially important for a program where compliance  is in large part
based on records kept by the CARs and other regulated parties.

      E.    The State's responsibility to monitor and assure the availability of
oxygenates
      Comments:  Some commenters stated that it is not the state's responsibility to
monitor and assure the availability of various oxygenates.  These commenters felt that
this function should be fulfilled by EPA.

      Response:  EPA agrees with the position stated in the Regulatory Negotiation
Agreement in Principal that the state should fulfill this function since these are state-
implemented programs and oxygenate-availability problems are likely to be localized.

      F.     Leadtime
      Comments:  Several  parties commented on the issue of whether EPA should
include an oxygenated gasoline availability leadtime at the terminals of 0, 5, or 30 days
in these guidelines.  Comments were received advocating all three time periods.
Commenters who did not support any leadtime at all tended to  believe that EPA
should rely on the workings of the market to ensure compliance at the beginning of
the control period.  One  commenter, who supported a 30-day leadtime, doubted that a
5-day leadtime would lead to conforming gasoline being sold at the start of the control
period.

      Response:   EPA believes that data collected to support the federal volatility
regulations supports a 5-day leadtime which should ensure that most retail stations will


                                      42

-------
be in compliance at the beginning of the control period.  The Agency notes that none
of the currently operating oxygenated gasoline programs utilize a leadtime.  Also, the
amount of oxygenates that would be needed to comply with a 30-day leadtime could
adversely impact the availability of oxygenates, especially in 1992. Finally, states could
decide to implement a longer leadtime should they deem that necessary.

      G.    Blending Allowance
      Comments:  One commenter suggested that the blending allowance should be
extended to terminals. Some commenters want the 0.2% allowance to be extended to
all ethers and alcohols limited by the 2.7% substantially similar ("sub sim") rule.  Some
commenters note that EPA should clearly distinguish between enforcement tolerance
and blending allowance.

      Response:   EPA agrees that the blending allowance should be available at all
points in the distribution system. EPA also agrees that the blending allowance should
be extended to all "sub sim" ethers and alcohols. EPA will address the issue of
enforcement tolerance in a subsequent action.

      H.    Availability of Nonoxvgenated Gasoline
      Comments:  Some parties expressed the opinion that there is not an
"availability  problem" for nonoxygenated gasoline to be blended with oxygenate
downstream from the refinery. These parties tended to believe that EPA should leave
the availability issue to market forces.  Some parties urge EPA to provide specific
guidance to states ensuring adequate levels of "clear" gasoline at terminals.

      Response:   EPA believes that the combination of market forces and  state
oversight wHI ensure the availability of nonoxygenated gasoline.

      I.     Registration Requirements
      Comments:  Commenters suggested various processing periods for CAR
registration. One commenter suggested that the processing period should be 60
days, but CARs should be allowed to register at a time later than the pre-control
period as long as those CARs refrain from sales until their registrations are processed.
Commenters supported processing  periods ranging from one to three months.  One
commenter suggested that individual States should determine how long a processing
period they need before the control  period in order to process CAR registrations.

      Response:   EPA has recommended a 30-day processing period. Mid-season
registrations are permitted under the guidelines. EPA encourages states to process
registration as  quickly as possible, and would not oppose a shorter processing period.

      J.    Reconciliation Period at the End of Averaging Period
      Comments:  Several parties suggested that the reconciliation period at the end
of the averaging period for balancing the books on the averaging standard should be

                                      43

-------
increased to 30 days.

      Response:   EPA has revised these guidelines to allow trading of credits to
occur for 30 days following the close of the averaging period.

      K.    Testing
      Comments:  Several commenters expressed support for ASTM D-4815 as a
widely used, dependable method. However, some commenters were concerned that
the ASTM method includes inadequate precision data. According to some
commenters, the method which has been  developed by EPA, the OFID  method, has
weaknesses including lack of evaluation, limited availability, and a poor service record.
Some commenters expressed concern over having to switch from one method to
another beginning in 1994 and suggested  that both methods be allowed.  One
commenter also expressed concern over the proposed quality control provisions.

      Response:   EPA believes that the ASTM D4815 method currently has the
shortcoming of poor precision (the ranges are too broad for enforcement  purposes)
and lacks the ability to quantify certain components expected to be allowed as
oxygenated blending agents, most notably ETBE and TAME.  However, ASTM is
currently working on a revision to this method, and it  is a procedure that many labs
currently have the capability to perform. Therefore, although the Agency prefers its
OFID method over the ASTM method at this time, the Agency agrees with the
commenters that it would be unreasonable to exclude ASTM 04815.

      EPA's OFID method does not have  any of the limitations associated with the
ASTM method at this time, and EPA feels the OFID to be more accurate and easier to
use than the ASTM method. When the OFID  method detects a peak, it  must be an
oxygenated compound and can therefore be incorporated in the  total oxygen
calculation directly. This is possible since  the OFID responds only to the percent
oxygen equivalent.  The ASTM method would simply record an uncalibrated peak as
an unknown, with no additional information as to the oxygen content of  the gasoline
blend.

      Another major reason for the Agency's preference for the OFID over the ASTM
D4815 is that the OFID test method will be calibrated to yield its results  in  terms of the
mass percent oxygenate contained in the fuel blend being analyzed, which is the unit
of measurement required by the statutory requirement. This.feature will obviate the
need for the rather complex calculations described above, necessary when converting
measurements from volume percent oxygenate to weight percent oxygen.

      EPA has knowledge of both methods.  Experience with the OFID method had
demonstrated it to be an extremely reliable instrument, requiring no service by the
distributor during the two years it has been operational in the EPA laboratory in Ann
Arbor, Michigan (NVFEL).  At this point, EPA has no reason to believe that the

                                     44

-------
instrument will require excessive maintenance.

      ASTM is also in the process of certifying an OFID method, and it is EPA's
intention to work with them in minimizing the differences between these procedures.
The Agency believes that the industry's concerns surrounding the OFID will be
reduced as exposure to the technique increases through the parallel efforts of ASTM.

      EPA has the task of minimizing the burden to all parties.  To this effect, the
Agency believes the least burden would be imposed by currently allowing both
methods as legal enforcement techniques.  At a later time, the Agency intends to
establish a federal testing tolerance for all oxygenate tests as mandated by the Act.
This tolerance will most likely be tied to the new OFID method, but the Agency will
continue to work with ASTM in order to allow industry the greatest flexibility in
purchasing and using  oxygenate laboratory testing  equipment.  This route to
enforcement will also allow those industries wishing to acquire an OFID in several
years 'to plan for and capitalize the expense. The use of  both methods will pose no
regulatory problems right now, since no bias exists between the two methods.

      It is the Agency's intention to provide the regulated parties with a description of
the quality control provisions that would be employed if the method were being .
performed by EPA personnel within the enforcement laboratory  at NVFEL  These
quality control provisions have been described as recommended practices.

      L.    CARs and Blender CARs
      Comments:  One commenter requested that EPA identify who the CAR is when
ethanol  blending occurs at the terminal.  The commenter notes  that there seems to be
a contradiction in the definitions  that suggests that a terminal which conducts blending
cannot be considered  a CAR but could be a blender CAR.  This commenter suggests
that eliminating the distinction between CAR and blender  CAR would clarify the
guidelines.  Also, the commenter notes that EPA should prohibit downstream parties
who are not registered CARs from changing the oxygen content of the gasoline.

      Response:   EPA has clarified the definitions and the responsibilities of the
CARs.   In particular, blending which results in a change in the oxygen content has
been prohibited by any party downstream of the terminal unless that party  is
registered as a CAR.

      Comments:  Some terminals do not take ownership  of product.  Many
commenters agreed that the CAR should be the gasoline owner.  Non-owners who
simply transfer product to trucks should be able to  rely on CAR, purchaser, or carrier
documentation. Some commenters thought that the requirement on CARs to assure
that oxygenated gasoline, once accounted for, is sold or  dispensed in the proper
control area may be unreasonable.  One commenter suggested that CARs should be
allowed  to fulfill the requirements by clearly indicating the control area for which the

                                      45

-------
gasoline was intended on the bill of lading.  Another commenter suggested that the
purchaser should be required to provide the CAR and terminal owner with a
certification of the destination of the gasoline.  The terminal owner should be allowed
to rely on that certification.

      Response:   EPA agrees that a terminal operator or CAR should be able to rely
on the representations of a purchaser as to the ultimate destination of the gasoline.
The terminal operator or CAR is required to indicate the destination, that is, control
area or non-control  area, on the bill of lading.  The terminal operator or CAR would not
be able to rely on this  representation, however, if he had information indicating that the
representation was false.

      Comments:   Some commenters thought that owners of gasoline at terminals
should be allowed to assign CAR responsibility to other parties through contractual
agreements.

      Response:   EPA agrees with this comment and has made the appropriate
changes to the guidelines. A party which owns gasoline  (for example, through an
exchange agreement) but wishes to reassign its CAR responsibilities (i.e.
recordkeeping, reporting, averaging, etc.) to another party better-suited to act as the
CAR for the gasoline in question could  contractually transfer the gasoline to the
assigned CAR under these guidelines.  Thus the assigning  CAR would have a zero
volume of gasoline to report.

      Comments:   Some commenters thought that terminals operators should  have
the option of using alternative methods, including LIFO (last in - first out) and FIFO
(first in - first out), to the running weighted average which appears in the guidelines.

      Response:   EPA believes that with the use of a running weighted average the
information associated with each batch of gasoline will be more accurate than it would
be with the use of the alternative suggested methods.  Also, the alternative methods
would make auditing by the states, internal auditors and independent CPAs more
difficult and costly.

      M.    SIP Submittal Dates
      Comments:   A few commenters felt that EPA should set a date before
November 1, 1992 for submission of SIP revisions. They supported the use of June 1,
1992. One party suggested that SIP revisions should be due on November 15,, 1992,
when other CO submittals are due.

      Response:   EPA has indicated  that due date is November 15, 1992. An earlier
due date wouMJ not  likely result in more assurance that states would implement the
oxygenated gasoline program in a timely manner since few states would be able to
comply with the earlier date because of the time their regulatory processes take.  Also,

                                       46

-------
in order to set a due date prior to November 1992, EPA would likely be required to
use notice and comment rulemaking.

      N.    Oxvqen Content
      Comments:  In response to the Agency's approach to oxygen content
conversions, commenters indicated no objections to the basic approach.  However,
commenters did suggest some slight changes.  Commenters indicated that, in Table 1
ETBE and TAME should have the same oxygen content.  It was also pointed out that
the table indicated "weight percent oxygen" when, in fact,  the values listed are "weight
fractions." Some commenters pointed out that the specific gravities in Table 1 should
be consistent with those being used by the American Society of Testing and Materials
(ASTM).

      Response:  The typographical  error has been corrected so that both ETBE and
TAME now show an oxygen weight or mass concentration of 0.1566. EPA agrees that
Table 1 should be consistent with the values used by ASTM.

      Comments:  One commenter suggested that the Agency continue its policy of
allowing up to two percent of unintentionally added MTBE in base gasoline used for
ethanol blending.20 The commenter further suggested that this policy be applied to
other oxygenates as well.

      Response:  The Agency has previously indicated that its policy allowing the
unintentional existence of no more than two percent MTBE in base gasoline also does
apply to waivered methanol blends (such as the DuPont waiver and the Texas
Methanol waiver21). Other oxygenates, blended at levels up to 2.7 percent oxygen
under the Agency's "substantially similar" definition,22 may  contain any mix of
alcohols or ethers (other than methanol). Hence, an allowance for inadvertent
"contamination" of one oxygenate blend with other ethers  or alcohols is not needed.

      Comments:  The Agency proposed that terminal blenders be given the option to
either use average specific gravities or to use actual measured specific gravities when
calculating oxygen content.  One commenter suggested that a terminal blender be
   20 EPA has previously indicated through the "sub sim" interpretive rule that it would
not consider it to be a violation of the gasohol waiver if up to 10 percent (by volume)
ethanol were added to unleaded gasoline containing no more than two percent (by
volume) MTBE.  However, the MTBE in the base fuel must be present only as a result
of commingling during storage and transport and not purposefully added as an
additional component to the ethanol blend.

   21 See 51  FR 15064 (April 22, 1986) and 53 FR 3636 (February 2, 1988).

   22 See 58 FR 5352 (February 11, 1991).

                                     47

-------
 required to use the same method throughout a single averaging period.

       Response: EPA sees no reasons why such a requirement is needed.

       0.    Quality Assurance for Carriers
       Comments: Many commenters opposed the requirement that carriers perform
 quality assurance programs in order to provide a defense in the presumptive liability
 scheme  included in these guidelines.  According to the commenters, carriers are
 solely concerned with the shipping and handling of oxygenated gasoline; because they
 do not own the products the handle, they have 'ittle of no incentive to alter the
 quantity  or quality of individual batches of gasoline. The commenters believe that a
 quality assurance program conducted by carriers would increase the costs of an
 oxygenated gasoline  program dramatically.

       Response:  EPA agrees with the commenters, and has eliminated the
 requirement that carriers perform quality assurance programs as part of a liability
 defense. Because carriers have no ownership interest in the products they handle, it
 is the Agency's belief that carriers have  little  or no incentive to alter the quantity or
 quality of the  gasoline that they handle.  Additionally, the Agency feels that current
 industry practice will involve a 'number of tests and cross-checks on carriers. The
 parties that own the gasoline normally will perform their own quality assurance
 programs; in  many instances this will occur both before and after the product is
 handled  by the carrier.  Therefore,  in today's notice EPA has relieved carriers of the
 need to prove defense through a quality assurance program.  However, it is important
to remember  that if it is believed that a carrier has caused a violation under these
guidelines, it is the carrier's responsibility to prove its defense in order to avoid
prosecution.  The standards of presumptive liability shall remain in place throughout
the gasoline distribution network.
      P.     Presumptive Liability
      Comments: Several commenters were opposed to the presumptive liability
proposal. Another commenter was concerned because a large proportion of their
customers are independents.

      Response:  EPA's experience with presumptive  liability in the lead
contamination and volatility programs indicates that it is an effective enforcement tool.
and that greater quality assurance takes place as a result.

      Q.    Averaging Programs
      Comments: Some commenters stated that EPA should strongly encourage
states to adopt averaging programs. One commenter expressed the view that
 averaging programs help to neutralize the competitive advantage of large refiners, who
 possess MTBE production capacity.

       Response:  EPA has strongly encouraged states to adopt averaging programs


                                      48

-------
and will continue to do so.

      R-    Product Transfer Documentation
      Comments:  Some eommenters felt that a requirement to .identify the type of
oxygenate is unnecessary and interferes with gasoline fungibility.  One commenter
suggested that the documentation show either the oxygen content of the gasoline in
weight percent or the type and volume of the oxygenate in the gasoline.

      Response:  EPA believes that the transfer document requirements, as written,
are necessary to allow purchasers to ensure that the gasoline they are purchasing
complies with tht program requirements. These requirements will also allow
regulators to trace gasoline which is found to be in violation of state requirements.

      S.    Oxygenated Gasoline "Caps"
      Comments:  Some eommenters stated that caps on oxygen content are
discriminatory and do not allow for competition between oxygenates. Two states are
considering imposing maximum oxygen content limits which would preclude the use of
10% ethanol blends.  One commenter  notes that unless EPA forcefully rejects caps,
the determination of the ethanol industry to produce and supply fuel ethanol in
amounts previously committed to will be undermined and impeded.

      Response:  In a separate notice, the Agency addresses these comments and
proposes a finding under section 211(c)(4)(A)(i) that "no control" on the maximum
oxygen content of gasoline as*a component of a winter oxygenated gasoline program
"is necessary."

      T.    Oxygenate  Purity
      Comments:  A number of eommenters opposed the inclusion of oxygenate
purity in the recordkeeping and reporting sections. The eommenters feel that the
purity has little relevance to the program and should not be required.

      Response:  The Agency disagrees with these eommenters.  It is believed that
the inclusion of information on oxygenate purity will provide a valuable cross-check to
the states as they evaluate the compliance data submitted by regulated parties at the
close of every control period.  By knowing the purity of the oxygenate used in each
blend of averaged gasoline, the states will be able to verify the  accuracy of the
calculations within the reports submitted by  the GARs and blender CARs.
                                      49

-------
 Appendix - Oxygenated Gasoline Credit Programs

 (a)    Scope.  This Appendix applies to credit programs employed in state
 oxygenated gasoline programs under  § 211 (m) of the Clean Air Act,  as amended (the
 Act).

 (b)    Definitions.
       (1)    Averaging period - The period of time over which  all gasoline sold or
 dispensed for use in a control area by  any control area responsible party must comply
 with the average oxygen content standard.
       (2)    Blender control area responsible party (Blender CAR) -  A person who
 owns oxygenated gasoline which is sold or dispensed from a control area oxygenate
 blending facility.
       (3)    Carrier -  Any person who transports, stores or causes  the transportation
 or storage of gasoline at any point in the gasoline distribution network, without taking
 title to or otherwise having ownership of the gasoline and without altering the quality or
 quantity of the gasoline.
       (4)    Control area - A geographic area in which only gasoline under the
 oxygenated gasoline program may be  sold or dispensed, with boundaries determined
 in accordance with  § 211 (m) of the Act.1
       (5)    Control area oxygenate blending facility -  Any facility or truck at which
 the oxygen content by  weight of gasoline intended for use in any control area is
 altered in any manner other than combining two or more fuels complying with the
 2.0% oxygen by weight minimum requirement.  The quality or quantity of gasoline may
 not be otherwise altered, except through the addition of deposit-control additives.
       (6)    Control area responsible  party (CAR) - A person who owns oxygenated
 gasoline which  is sold or dispensed from a control area terminal.
       (7)    Control area terminal -  A terminal which is capable of receiving gasoline
 in bulk, i.e., by  pipeline, marine vessel  or barge, and/or at which gasoline is altered
 either in quantity or quality, excluding the addition of deposit control additives.
 Gasoline which is intended for  use in any control area is sold or dispensed into trucks
 at these control area terminals.
       (8)    Control period -  The period during  which oxygenated gasoline must be
 sold and dispensed in any control area, pursuant to § 211(m)(2) of the Act.2
       (9)    Distributor - Any person  who transports or stores or causes the
transportation or storage of gasoline at any point between any gasoline refinery or
 importer's facility and any retail outlet or wholesale purchaser-consumer's facility.
    1 The boundaries of the control areas are noted in a separate Federal Register
notice published today.

    2 EPA is required to determine the control periods, set by §211(m)(2) of the Act, as
that portion of the year in which the area is "prone to high ambient concentrations of
carbon monoxide."  In another Federal Register notice published today, EPA is
establishing lengths of the control periods for the different areas covered by §211(m).
                                    50

-------
      (10)  Gasoline -  Any fuel sold for use in motor vehicles and motor vehicle
engines and commonly or commercially known or sold as gasoline.
      (11)  Nonoxygenated gasoline - Any gasoline which does not meet the
definition of oxygenated  gasoline.
      (12)  Oxygen content of gasoline blends -  Percentage of oxygen by weight
contained in a gasoline blend, based upon its percentage oxygenate by volume,
excluding denaturants and other non-oxygen-containing components. All
measurements shall be adjusted to 60 degrees Fahrenheit.
      (13)  Oxygenate -  Any substance which, when added to gasoline, increases
the anhount of oxygen in that gasoline blend. Lawful use of any combinatior  of these
substances requires that they be "Substantially  Similar" under § 211.(f)(1) of the Clean
Air Act3 or be permitted under a waiver granted by the Administrator under the
authority of § 211(f)(4) of the Clean Air Act.
      (14)  Oxygenate blender - A person who owns, leases,  operates,  controls or
supervises a control area oxygenate blending facility.
      (15)  Oxygenated gasoline - Any gasoline which contains at least 2.0% oxygen
by weight and has been included in the oxygenated gasoline program accounting by a
control area responsible party and which is intended to be sold or dispensed for use
in any control area.
      (16)  Refiner - Any person who owns, leases, operates, controls, or
supervises a refinery which produces gasoline for use in a control area.
      (17)  Refinery  - A plant at which gasoline is produced.
      (18)  Reseller - Any person who purchases gasoline and resells or transfers it
to a retailer or a wholesale purchaser-consumer.
      (19)  Retail outlet -  Any establishment at which gasoline is sold or offered for
sale to the ultimate consumer for use in motor vehicles.
      (20)  Retailer - Any person who owns,  leases, operates, controls or
supervises a retail outlet.
      (21)  Terminal -  A facility at which  gasoline is sold,  or dispensed into trucks
for transportation  to retail outlets or wholesale purchaser-consumer facilities.
      (22)  Wholesale purchaser-consumer -  Any organization that is an ultimate
consumer of gasoline and which purchases or obtains gasoline from a supplier for use
in motor vehicles  and  receives delivery of that product into a storage tank of at least
550-gallon capacity substantially under the control  of that organization.

(c)    Average oxygen content standard.
      (1)   All gasoline sold or dispensed during the control period for use in each
control area by each CAR or blender CAR as defined in paragraph (b) of this
Appendix, shall be blended during each averaging period to contain an average
oxygen content of not less than 2.7% by weight.  Oxygen content calculations shall be
performed in accordance with paragraph (d).
      (2)   The  averaging period over which  all gasoline sold or dispensed in the
control area is to  be averaged shall be equal to the length of the control period as
     56 FR 5352 (February 11, 1991).


                                         51

-------
established by the Administrator, except that programs with control periods of six
months or longer shall have averaging periods of three months.4

(d)    Sampling, testing and oxygen content calculations.
       (1)    For the purpose of determining compliance with the requirements of this
Appendix, the oxygen content of gasoline shall be determined by:
             (i)     Sampling.  Use of the sampling methodologies specified in
Appendix A of this Appendix to obtain a representative sample of the gasoline to be
tested;
             (ii)     Testing.
                   (A)    Use of one of the test methods specified in Appendices B
and C of this Appendix.  These methods are used to determine the mass
concentration of each oxygenate in the gasoline sampled;  or
                   (B)    Use of any alternative test method which has been
successfully evaluated and approved by the Agency;  and
             (iii)    Oxygen Content Calculations.
                   (A)    Calculation of the oxygen content of the gasoline sampled
by multiplying the mass concentration of each oxygenate in the gasoline sampled by
the oxygen molecular weight contribution of the oxygenate set forth in paragraph
(d)(2) of this Appendix; and
                   (B)    All volume measurements shall be adjusted-to 60 degrees
Fahrenheit.
       (2)    For purposes of this Appendix, the oxygen  molecular weight
contributions  of oxygenates currently approved for  use in the United States are the
following:
    4 EPA shall determine the length of the control period during the State
Implementation Plan review and approval process.
                                  52

-------
Table 1.  Specific Gravity and Weight Percent Oxygen of Common Oxygenates
Oxygenate
methyl alcohol
ethyl alcohol
normal propyl alcohol
isopropyl alcohol
normal butyl alcohol
isobutyl alcohol
secondary butyl alcohol
tertiary butyl alcohol
methyl tertiary butyl ether (MTBE)
tertiary amyl methyl ether (TAME)
diiso propyl ether (DIPE)
ethyl tertiary butyl ether (ETBE)
Weight fraction
oxygen
0.4993
0.3473
0.2662
0.2662
0.2158
0.2158
0.2158
0.2158
0.1815
0.1566
0.1566
0.1566
Specific Gravity
at 60 deg F
0.7963
0.7939
0.8080
0.7899
0.8137
0.8058
0.8114
0.7922
0.7460
0.7752
0.7300
0.7452
(e)  Alternative compliance options.  Each CAR or blender CAR shall comply with the
standard specified in paragraph (c) of this Appendix by means of the method set forth
in either paragraph (e)(1) or (e)(2) of this Appendix.
      (1)   Compliance calculation on average basis.
            (i)    To determine compliance with the standard in paragraph (c), the
CAR or blender CAR shall, for each averaging period and for each control area:
                  (A)   Calculate the total volume of gasoline sold or dispensed for
use in the control area which is the sum of:
                        (1)   The volume of each separate batch or truckload of
oxygenated gasoline that is- sold or dispensed;
                        (2)    Minus the volume of each separate batch or
truckload of oxygenated gasoline that is sold or dispensed for  use in a different control
area;
                        (3)    Minus the volume of each separate batch or
truckload of oxygenated gasoline that is sold or dispensed for  use in any non-control
area.
                  (B)   Calculate the required total oxygen credit units.  Multiply the
total volume in gallons of oxygenated gasoline sold or dispensed for use in the control
area (as  determined  by (e)(1)(A) above) by 2.7 percent.
                  (C)  Calculate the actual total oxygen credit units generated.
The actual total oxygen credit units generated is the sum of the volume of each batch
or truckload of oxygenated  gasoline that was sold or dispensed for use in the control
                                  53

-------
 area (as determined by (e)(1)(A) above) multiplied by the actual oxygen content by
 weight associated with each batch or truckload.
                   (D)   Calculate the adjusted actual total oxygen credit units. The
 adjusted actual total oxygen content units is the sum of the actual total oxygen credit
 units generated (as determined in (e)(1)(C) above);
                         (1)    Plus the total  oxygen  credit units purchased or
 acquired through trade; and
                         (2)    Minus the total oxygen credit units sold or given
 away through trade.
                   (E)    Compare the adjusted actual total oxygen credit units with
 the required total oxygen credit units.  If the adjusted actual total content oxygen credit
 units is greater'than or equal to the required total oxygen credit units, then the
 standard in paragraph (c) is met.  If the adjusted actual total oxygen credit units is less
 than the required total oxygen  credit units then the purchase of oxygen credit units is
 required in order to achieve compliance.
                   (F)    In transferring oxygen credit units, the transferor shall
 provide the transferee with the volume and oxygen content by weight of the gasoline
 associated with the  credits.
             (ii)     To determine the oxygen credit units associated with each batch
 or truck load of oxygenated gasoline sold or dispensed into the control area, use the
 running weighted oxygen content (RWOC) (see (iii) below) of the tank from which the
 batch or truckload was received at the time the batch or truckload was received.  In
 the case of batches or truckloads of gasoline to which  oxygenate is added outside of
 the terminal storage tank  from  which it was received, use the weighted average of  the
 RWOC and the oxygen content added as a result of the volume of the additional
 oxygenate added.
             (iii)     Running weighted oxygen content (RWOC).  The RWOC accounts
 for the volume and oxygen content of all gasoline which enters or leaves the terminal   .
 storage tank, and all oxygenates which  are added to the tank.  The RWOC must be
 calculated each time gasoline enters or leaves the tank or whenever oxygenates are
 added to the  tank. The RWOC is calculated weighing the following:
                   (A)    The volume and oxygen content of the gasoline in the
storage tank at the beginning of the averaging period;
                   (B)    The volume and oxygen content by weight of gasoline
entering the storage tank;
                   (C)    The volume and oxygen content by weight of gasoline
 leaving the storage tank;  and
                   (D)    The volume,  type and oxygen content by weight of the
oxygenates added to the  storage tank.
             (iv)    Credit transfers. Credits may be used in the compliance
calculation in (e)(1)(i)(A),  provided that:
                   (A)    The credits are generated in the same control area  as they
are used, i.e., no credits may be transferred between nonattainment areas;
                   (B.)    The credits are generated in the same averaging period as
they are used;
                   (C)    The ownership of credits is transferred only between CARs
                                     54

-------
 or Blender CARs;
                  . (D)   The credit transfer agreement is made no later than 30
 working days after the final day of the averaging period in  which the credits are
 generated; and
                   (E)   The credits are properly created.
             (v)    Improperly created credits.
                   (A)   No party may transfer any credits to the extent such a
 transfer would result in the transferor having a negative credit balance at the
 conclusion of the averaging period for which the credits were transferred.  Any credits
 transferred in violation of this paragraph are improperly created  credits.
                   (B)   In the case of credits which were  improperly created, the
 following provisions apply:
                         (1)   Improperly created credits  may not be used,
 regardless of a credit transferee's  good faith belief that it was receiving valid credits;
                         (2)   The transfer of credits in  violation of (A)  above
 constitutes a violation of these requirements, for which the transferor will be deemed to
 be in violation; and
                         (3)   Where any credits are transferred in violation of (A)
 above, the transferor's properly-created credits will be applied first to any credit
 transfers before the transferor may apply any credits to achieve its  own compliance.
                         (4)   Where any credits are transferred in violation of (A)
 above, the transferor shall be held legally and financially liable for any penalties or
 damages incurred by the transferee as a result of the invalid transaction.
      (2)     Compliance calculation on per-gallon basis.   Each  gallon of gasoline sold
 or dispensed by a CAR or Blender CAR for use within each control area during the
 averaging period as defined in paragraph (c) shall have an oxygen  content of at least
 2.7% by weight.  In addition, the CAR or Blender CAR is prohibited from selling
 oxygen  credits based on gasoline  for which compliance is  calculated under this
 alternative per-gallon method.

 (f)    Minimum oxygen content.
      (1)     Any gasoline which is sold or dispensed by a CAR or a Blender  CAR for
 use within a control area, as defined in paragraph (b), during the control period shall
 contain  not less than 2.0% oxygen by weight unless it is sold or dispensed to another
 registered CAR or Blender CAR. This requirement shall begin five days before the
 applicable control period and shall apply until the end of that period.
      (2)     This requirement shall apply to all parties downstream of the CAR. Any
gasoline which is offered for sale, sold or dispensed to an  ultimate  consumer  within a
 control area, as defined in paragraph (b), shall contain not less than 2.0% oxygen by
weight.  This requirement shall apply during the entire applicable control period.
      (3)     Every refiner or importer must determine the  oxygen content of each
 batch of gasoline produced by use of one of the methodologies in  the Appendices as
described in section  (d). This determination shall include the percent oxygen by
weight,  the type of oxygenate and  the oxygenate percent by volume.

 (g)    Registration.
                                   55

-------
      (1)   At least one month in advance of eny control period in which a party will
meet the definition of CAR or Blender CAR, such party shall petition for registration as
a CAR or Blender CAR in each state that the party intends to serve. A party may
petition for registration as a CAR or Blender CAR after the beginning of a control
period but should do so at least 30 days before  it plans to begin conducting activities
as a CAR or Blender CAR.  This petition for registration shall be on forms prescribed
by the state and shall include the following information:
            (i)     The name and business address of the control area responsible
party;
            (ii)     The address and physical location of each of the control area
terminals from which the control area responsible party operates;
            (iii)    The address and physical location of each control area oxygenate
blender facility which is owned, leased,  operated, controlled or supervised by a
Blender CAR; and
            (iv)    The address and physical location where documents which are
required to be retained by this Appendix will be kept by the CAR.
      (2)   Within thirty days of any occasion when the registration information
previously supplied  by a CAR becomes incomplete or inaccurate, the CAR or Blender
CAR shall submit updated registration information to the state.
      (3)   No party shall participate in the averaging program  under paragraph  (e)
of this Appendix as  a CAR or Blender CAR until  it has been notified  by the state that it
has been registered as a CAR or Blender CAR and  has been issued a unique CAR
identification number.  This should occur within 30 days of the submission of the
registration application to the state.  Registration by a  state shall  be  valid for the time
period specified by the state. The state shall issue each CAR and Blender CAR a
unique identification number.

(h)    Recordkeeping and reporting.
      (1)    Records.  All parties in the gasoline distribution network as described
below shall maintain records containing compliance information enumerated or
described below.  These records shall be retained by the regulated parties for a period
of time established by the state  consistent with its relevant statute of limitations.
            (i)     Refiners and Importers.  Refiners and importers shall, for each
separate quantity of gasoline produced or imported for use in a control area during a
control period, maintain records containing the following information:
                   (A)   Results of the tests performed to determine the types of
oxygenates and percentage  by volume;
                   (B)   Oxygenate content by volume;
                   (C)   Oxygen content by weight;
                   (D)   Total volume of gasoline; and
                   (E)   Name and address of the party to whom each separate
quantity of gasoline was sold or transferred.
            (ii)     Control area terminal operators and CARs. Persons who own,
lease, operate or control gasoline terminals which serve control areas, or where
appropriate, any CAR truck- or terminal-lessee who subleases any portion, of a leased
or tank or terminal to other persons, and  all CARs shall maintain records containing
                                    56

-------
the following information:                •
                   (A)   The owner of each batch of gasoline handled by each
regulated facility if known, or the storage customer of record;
                   (B)   Volume of each batch or truckload of gasoline going into or
out of the terminal;
                   (C)   For all batches or truckloads of gasoline leaving the
terminal, the RWOC of the batch or truckload;
                   (D)   Type of oxygenate, purity, and percentage by volume if
available;                .                           .
     -              (E)   Oxygen content by weight of all batches or truckloads
received at the terminal;
                   (F)   Whether the gasoline is intended for use within a control
area or not;
                   (G)   The destination of each tank truck sale or batch of gasoline
as furnished in a written declaration by the purchaser of the gasoline;
                   (H)   The name and address of the party to whom the gasoline
was sold or transferred and the date of the sale or transfer; and
                   (I)    Results of the tests for oxygenates,  if performed, for each
sale or transfer, and who performed the tests.
            (iii)   CARs and Blender CARs. CARS and Blender CARs must maintain
records containing the information listed in paragraph (ii) above, plus the following
information:
                  (A)   CAR or Blender CAR identification number;
                  (B)   Records supporting and demonstrating compliance with the
averaging standard listed in paragraph (c) of this Appendix;
                  (C)   For any credits bought, sold, traded or transferred, the
dates of the transactions, the names, addresses and CAR or Blender CAR numbers of
the CARs or Blender CARs involved in the individual transactions, and the amount of
credits (oxygen content and volume of gasoline) transferred. Any credits transferred
must be accompanied by a demonstration of how those credits were calculated.  Also
included must  be adequate documentation that both parties have agreed to all credit
transactions within 30 working days following the close of the relevant averaging
period;
                  (D)   The name and address of the auditor, and the results of the
attestation engagement conducted pursuant to paragraph (j) of this Appendix;
                  (E)   The name and address of the party from which each
shipment of gasoline was received, and the date when it was received;
                  (F)   Data on each shipment of gasoline  received, including:
                        (1)   The total volume of each shipment;
                        (2)   Type of oxygenate, purity, and percentage by
volume; and
                        (3)   Oxygen content by weight;
                  (G)   The volume of each receipt of bulk  oxygenates;
                  (H)   The name and address of the parties from whom bulk
oxygenate was received;
                  (I)    Date and exact destination,  if available, of each sale of
                                  57

-------
gasoline;
                  (J)    Whether the gasoline was destined for a control area or
not;
                  (K)    Data on each shipment of gasoline sold or dispensed
including;
                        (1)    The volume of each shipment;
                        (2)    Type of oxygenate, purity, and percentage by
volume; and
                        (3)    Oxygen content by weighty
  ~                (L)    Documentation of the results of all tests performed
regarding the oxygen content of gasoline; and
                  (M)   The names, addresses and CAR or Blender CAR
identification numbers of the parties to whom any gasoline was sold or dispensed, and
the dates of these transactions
            (iv)   Retailers and wholesale purchaser-consumers within a control
area must maintain the following records:
                  (A)    The names, addresses and CAR or Blender CAR
identification numbers of the parties from whom all shipments of gasoline were
purchased or received, and the dates on which they were received; and
                  (B)    Data on every shipment of gasoline bought, sold or
transported, including:
                        (1)    Total volume of each  shipment;
                        (2)    Type of oxygenate, purity, and percentage by
volume;
                        (3)    Oxygen content by weight; and
                        (4)    Whether the gasoline is intended for use in a control
area or not.
      (2)    Reports.
            (i)    Each CAR and Blender CAR shall submit a self-audit report for
each averaging period as defined in paragraph (c) reflecting the compliance
information detailed in paragraph (e) of this Appendix.  Reports are due on the 30th
day of each month following the averaging period for which the information is required.
These reports shall be filed using forms provided by the state.
            (ii)    CARs or Blender CARs shall also submit attestation engagement
reports as required by paragraph (j) of this Appendix.  Attestation engagements are to
be conducted at the end of the control period,  or every 6 months, whichever is
shorter. The report is to be submitted to the state within 120 days following the end of
the pericd covered by the engagement.
      (3)    Transfer Documents. Each time that physical custody or title of gasoline
destined for a control area changes hands other than when gasoline is sold or
dispensed for .use in motor vehicles at a retail outlet or wholesale purchaser-consumer
facility, the trafflferor shall provide to the transferee, in addition to,  or as part of,
normal bills of lading, invoices,  etc., a document containing  information on that
shipment.  This document shall accompany every shipment of gasoline to a control
area after it.has been dispensed by a terminal, or the information shall be included in
the normal paperwork which accompanies every shipment of gasoline.  The
                                    58

-------
information shall legibly and conspicuously contain the following information:
             (i)     The date of the transfer;
             (ii)    The name, address, and CAR or Blender CAR identification
number, if applicable, of the transferor;
             (iii)    The name, address and CAR or Blender CAR identification
number, if applicable, of the transferee;
             (iv)    The volume of gasoline which is being transferred;
             (v)    The proper identification of the gasoline as nonoxygenated or
oxygenated;
             (vi)    The location of the gasoline at the time of the transfer;
             (vii)   Type of oxygenate; and
             (viii)   For gasoline which is in the gasoline distribution network between
the refinery or import facility and the control area terminal, the oxygen content by
weight and the oxygenate volume of the gasoline.

(i)    Prohibited activities.
      (1)    During the control period, no refiner, importer, oxygenate blender, carrier,
distributor or reseller may manufacture, sell, offer for sale, dispense, supply, offer for
supply,  store, transport, or cause the transportation of:
             (i)     Gasoline which contains less than 2.0% oxygen by weight, for use
during the control period in a CO nonattainment area  subject to the requirements of §
211(m)  of the Act; or.
             (ii)    Gasoline represented as oxygenated which has an oxygen content
which is improperly stated in the documents which accompany such gasoline.
      (2)    No retailer or wholesale purchaser-consumer may dispense, offer for
sale, sell or store, for use during the control period, gasoline which contains less than
2.0% oxygen by weight in a CO nonattainment area subject to the requirements of §
211 (m)  of the Act.
      (3)    No party may operate as a CAR or Blender CAR or represent itself as
such unless it has been properly registered by the state(s) involved. No CAR or
Blender CAR may offer for sale, store, sell or dispense gasoline to any person not
registered as a CAR for use in a control area, unless:
        '~     (i)     The average oxygen content of the gasoline during the averaging
period meets the standard established in  paragraph (c) of this Appendix; and
             (ii)    The gasoline contains at least 2.0% oxygen by weight  on a per-
gallon basis.
      (4)    For terminals which sell or dispense gasoline intended for use in a
control area during the control period, the terminal owner or operator  may not accept
gasoline into the terminal unless:
             (i)     Transfer documentation accompanies it containing the information
specified in paragraph  (h)(3);.and
             (ii)    The terminal owner or operator conducts a quality assurance
program to verify the accuracy of this information.
      (5)    No person may sell or dispense  nonoxygenated gasoline for  use in any
control area during the control period, unless:
             (i)     The nonoxygenated gasoline is segregated from oxygenated
gasoline;
                                       59

-------
             (ii)    Clearly-marked documents accompany the nonoxygenated
gasoline marking it as "nonoxygenated gasoline, not for sale to ultimate consumer in a
control area," and
             (iii)    The nonoxygenated gasoline is in fact not sold or dispensed to
ultimate consumers, during the control period, in the control area.
       (6)    No named party may fail to comply with the recordkeeping and reporting
requirements contained in section (h).
       (7)    No person may sell, dispense or transfer oxygenated gasoline, except for
use by the ultimate consumer at a retail outlet or wholesale purchaser-consumer
facility, without transfer documents which accurately contain the information required
by section (h)(3).          /
       (8)    Unless registered  as a CAR or Blender CAR, no person downstream of a
terminal may combine gasoline complying with the 2.0% oxygen by weight minimum
with gasoline not complying with the 2.0% minimum requirement if that gasoline is to
be sold or dispensed in a control area during a control period.
       (9)  liability for violations of the prohibited  activities. .
             (i)     Where the gasoline contained in any storage tank at any facility
owned, leased, operated,  controlled or supervised by any retailer, wholesale
purchaser-consumer, distributor, reseller, carrier, refiner, importer, or oxygenate
blender is found in violation  of the prohibitions described in sections (1)(i)  or (2) of this
paragraph, the following persons shall be deemed in violation:
                   (A)    The retailer, wholesale purchaser-consumer,  distributor,
reseller, carrier, refiner, importer, or oxygenate blender who owns, leases, operates,
controls or supervises the facility where the violation is found; and
                   (B)    Each oxygenate blender, distributor, reseller, and carrier
who, downstream of the control area terminal, sold, offered for sale,  dispensed,
supplied, offered for supply, stored, transported, or caused the transportation of any
gasoline which is in the storage tank containing gasoline found to be in violation.
             (ii)     Where the gasoline contained in any storage tank at any facility
owned, leased, operated,  controlled or supervised by any distributor, reseller, carrier,
refiner, importer, or oxygenate blender is found in  violation of the prohibitions
described in section (1)(ii) or (2) of this paragraph, the following persons shall be
deemed in violation:
                   (A)    The retailer, wholesale purchaser-consumer,  distributor,
reseller, carrier, refiner, importer, or oxygenate blender who owns, leases, operates,
controls or supervises the facility where the violation is found; and
                   (B)    Each refiner, importer, oxygenate blender, distributor, and  .
reseller who manufactured, imported, sold, offered for sale, dispensed, supplied,
offered for supply, stored, transported, or caused  the transportation  of any gasoline
which is in the storage tank  containing gasoline found to be in violation.
       (10)   Defenses for prohibited activities.
             (i)     In any case in which a refiner, importer, CAR, distributor or reseller
would be in violation under paragraph (i)(1), it shall be deemed not in violation if it can
demonstrate:
                   (A)    That the violation was not caused by the regulated party or
its employee  or agent;
                   (B)    That it possesses transfer documents which  support the

-------
oxygen content of all gasoline in its possession. These transfer documents are
required to accompany all oxygenated gasoline in the gasoline distribution network
under paragraph (h).
                   (C)         A quality assurance sampling and testing program
carried out by the regulated party, as described in (i)(11).
             (ii)     In any case in which a carrier would be in violation under
paragraph (i)(1), it shall be deemed not in violation if it can demonstrate:
                   (A)    That the violation was not caused by the regulated party or
its employee or agent;
                   (B)    That it possesses product transfer documents  required
under, paragraph (h) for all the gasoline in its possession,  and that the oxygen
contents reflected in these documents are consistent with  the oxygen content of such
gasoline.
             (iii)    In any case in which a retailer or wholesale purchaser-consumer
would be in  violation under paragraph (i)(2). it shall be deemed not in violation if it can
demonstrate:
                   (A)    That the violation was not caused by the regulated party or
its employee or agent; and
                   (B)    That it possesses transfer documents which support the
oxygen content of all gasoline in  its possession. These transfer documents are
required to accompany all oxygenated gasoline in the gasoline distribution network
under paragraph  (h).
             (iv)    Where a violation is found at a facility which is operating under the
corporate, trade  or brand name of a refiner, that refiner must show, in addition to the
defense elements required by paragraph (i)(1C)(i), that the violation was caused by:
                   (A)    An act in violation of law (other  than the Act or this part) or
an act of sabotage or  vandalism; or
                   (B)    The action of any reseller, distributor, oxygenate blender,
carrier, or a  retailer or wholesale  purchaser-consumer which is supplied by any of the
persons listed above in paragraph (i)(10)(i),  in violation of a contractual undertaking
imposed by  the refiner designed  to prevent such action, and despite periodic sampling
and testing by the refiner to ensure compliance with such  contractual obligation; or
                   (C)    The action of any carrier or other  distributor not subject to a
contract with the refiner but engaged  by the refiner for transportation of gasoline,
despite specification or inspection of procedures and equipment by the refiner or
periodic sampling and testing which are reasonably calculated to prevent  such action.
             (v)     In this paragraph (i)(10), the term "was caused" means that the
party  must demonstrate by reasonably specific showings,  by direct or circumstantial
evidence, that the violation was caused or must have been caused by another.
       (11)   Quality Assurance Program. In order to demonstrate an acceptable
quality assurance program, any party except for a carrier,  wholesale purchaser-
consumer or retailer must conduct itself or commission an independent testing service
to conduct pelfodic sampling and testing to determine  if the  oxygenated gasoline has
oxygen content which is consistent with the product transfer documentation.

G)    Attestation engagements.
       (1)    The attestation engagement shall consist of performing the agreed-upon
                                        61

-------
 procedures set forth in the guidelines in accordance with either the Codification of
 Standards for the Professional Practice of Internal Auditing as promulgated by the
 Institute of Internal Auditors, Inc. or the American Institute of Certified Public
 Accountants' (AlCPA's) Statements on Standards for Attestation Engagements and
 using statistical sample design parameters provided by the Agency.5
       (2)    The attestation engagement shall be conducted by either an internal
 auditor employed by the CAR or a Certified Public Accountant (CPA).  This attestation
 engagement may be performed by either an internal auditor employed  by the CAR or,
 if a CAR has no internal auditors available, an independent CPA or a firm of
 independent CPAs.  Internal auditors must be either Certified Internal Auditors (CIAs)
 or members in good standing of the Institute of Internal Auditors, Inc (IIA).
       (3)    The CPA is required to comply with the general code of conduct and
 ethics  as prescribed by the state in which he or she is licensed and, if applicable, a
 member in good standing  of either the AICPA or the IIA.
       (4)    The attestation engagement shall include the following agreed-upon
 procedures, as appropriate, for the CAR'S standardized reporting form(s):
             (i)     Read the report completed by management and filed with the
 state agency.
             (ii)     Obtain from the CAR an inventory  reconciliation summarizing
 receipts and deliveries of all gasoline, gasoline blendstocks, and oxygenates for CARs
 serving a control area.
                   (A)    Test mathematical accuracy of inventory reconciliation.
                   (B)    Agree beginning and ending inventory amounts to
 company's perpetual inventory records.
                   (C)    Agree deliveries into the control area to state report, if
    5 In performing the attest.engagement, the internal auditor or CPA shall determine
the sample size for each population according to the following table:

            Number in Population (N)             Sample Size

                  66 or larger                         59

                  41-65                            41

                  26-40                            31

                  0-25                      N or 24, whichever is
                                                      smaller

     ' The number of populations from which samples should be drawn will vary
depending on the circumstances.  Sample.items should be selected in such a way that
the sample can "be expected to be representative of the population.
      If the CPA agrees to use some other form of sample selection and some other
method to determine the sample  size, that agreement should be summarized in the
CPA's report.
                                     62

-------
applicable..
             (iii)    Obtain listing of all gasoline, gasoline blendstocks, and oxygenate
receipts during the period.
                   (A)    Test mathematical accuracy of listing.
                   (B)    Agree amounts to inventory reconciliation.
                   (C)    Select a representative sample of individual receipts of
gasoline, gasoline blendstocks, and oxygenates and trace details back to source
documents.
             (iv)    Obtain listing of all gasoline, gasoline blendstocks, and oxygenates
sold or dispensed during the period.
                   (A)    Test mathematical accuracy of listing.
                   (B)    Agree amounts to inventory reconciliation report.
                   (C)    Select a representative sample of individual batches sold or
dispensed both into and outside the control area.
                         (1)    Agree volumes for the sample items to original bill of
lading  or other source documents.
                         (2)    For sales or deliveries into the control area,
determine that oxygenate content is at least two percent by examining bills of lading.
             (v)    Using the volume of oxygenated gasoline sold or dispensed into
the control area from the inventory reconciliation report, recalculate the number of
oxygen content units required by multiplying volume by 2.7% and agree to state
report.
             (vi)    Recalculate the actual total oxygen credit units generated by
adding the oxygen content units of each batch or truckload of oxygenated gasoline
that was  sold or dispensed in the control area as determined in section (v) above.
These  units are generated by multiplying the actual oxygen content by weight
associated with each batch or truckload by the volume.
             (vii)    Recalculate the adjusted actual total oxygen credit units as follows:
                   (A)    The actual total oxygen credit units generated from section
(vi);
                   (B)    Plus the total oxygen credit units purchased or acquired
through trade; and
                   (C)    Minus the total  oxygen credit units sold or given away
through trade.
             (viii)   The following steps apply to the testing of the actual total oxygen
content from  section (vi) and are applicable based on method of blending:
                   (A)    For CARs using rack- and splash-blending, recompute
oxygen content by weight for a representative sample of deliveries based on detailed
meter  readings of gasoline, blendstocks and oxygenate receipts.
                   (B)    For CARs using in-tank blending of gasoline, blendstocks
and oxygenates, obtain register of running weighted oxygen content by tank and:
                         (1)   Using the individual sample items from sections (iii)
and (iv) above, test calculation of running  totals.
                         (2)   Where laboratory analysis is used within the CAR'S
weighted average calculation, select individual analysis reports of oxygenated gasoline
receipts and deliveries during the period on a representative sample basis.
                               (a)   Review laboratory results for consistency with
                                  63

-------
.CAR'S calculations noting oxygen volume and specific gravity.
                               (b)    Recalculate oxygen by weight.
                               (c)    Agree information on lab reports to underlying
delivery and receiving documentation.
             (ix)    Obtain register of oxygen credit unit purchases and sales and
select separate representative samples of individual purchased credits and individual
sales credits.
                   (A)   Agree selected credit unit transactions to the underlying
contract and/or other supporting documentation noting specific volumes and oxygen
content of the gasoline associated with the credits.
                   (B)   Agree to the underlying contract and/or supporting
documentation that the credits are generated in the same control areas as they are
used (i.e., no credits may be transferred between nonattainment areas).
                   (C)   Agree to the underlying contract and/or supporting
documentation that the credits are generated in the same averaging period as they are
used                                                      •
                   (D)   Agree to the underlying contract and/or supporting
documentation that the ownership of credits is transferred only between CARs.
                   (E)   Agree to the underlying contract and /or supporting
documentation that the credit transfer agreement was made no later than 30  working
days after the final day of the averaging period in which the credits are generated.
             (x)    Prepare a report to CAR  management in accordance with the
Codification of Standards for the Professional Practice of Internal Auditing as
promulgated by the HA or the AlCPA's Statements on Standards for Attestation
Engagements indicating the results of performing the above procedures.   This report
should include, in addition to the information  described in Appendix D, a declaration of
the internal auditor's or the CPA's professional  credentials.
                                             64

-------
                             Appendix A -- Sampling Procedures
            EPA's sampling procedures are detailed in Appendix D of'40 CFR 80.

                               Appendix B - Testing Procedure

             PROPOSED REVISION OF ASTM DESIGNATION: D^ai5-89

 Standard Test Method for
 Determination of MTBE, ETBE, TAME, DIPE, tertiary-Amy! Alcohol and Q to C4 Alcohols
 in Gasoline by Gas Chromatographyl

 This standard is issued under the fixed designation D-4815; the number immediately following
 the designation indicates the year of original adoption or, in the case of revision, the year of
 last revision. A number in parentheses indicates the year of last reapproval. A superscript
 epsilon (e) indicates an editorial change since the last revision or reapproval.

 ^This test method is. under the jurisdiction of ASTM Committee D-2 on hydrocarbons and is
 the direct responsibility of Subcommittee D.02.04 on methods of analysis.

This test method supercedes ASTM D-4815-89.2 This test method has undergone extensive
resivion in order to address current requirements, for the determination of oxygenates in fuels.

1.     Scope
      1.1   This test method is designed for the determination of ethers and alcohols in
            gasolines by gas chromatography. Specific compounds determined are: methyl
            tert-butylether (MTBE), ethyl tert-butylether (ETBE), tert-amvlmethvlether
            (TAME), diisopropylether (DIPE), methanol, ethanol, isopropanol, n-propanol,
            isobutanol, tert-butanol. sec-butanol, n-butanol, and tert-pentanol (tert-
            amylalcohol).

      1.2   Individual ethers are determined from 0.1 to 20.0 mass%. Individual alcohols are
            determined from 0.1 to 12.0 mass%. Equations used to convert to mass%
            oxygen and to volume% of individual compounds are provided.

      1.3    SI (metric) units are preferred and used throughout this standard. Alternate
            •units, in common usage, are also provided to increase clarity and aid the users of
            this method.

      1.4    Alcohol-based fuels such as M-85 and E-85, MTBE product, ethanol product
            and denatured alcohol are specifically excluded from this method. The
            methanol content of M-&5 fuel is considered beyond the operating range of the
            system.
                                           65

-------
       1.5    Benzene, while detected, cannot be quantified using this test method and must
              be analyzed by alternate methodology (Test Methods D-3606 or D-4420).2

       1.6    This standard does not purport to address all of the safety problems associated with
              its use. It is the responsibility of the user of this standard to establish appropriate
              safety and health practices and determine the applicability of regulatory limitations
             prior to use.

 2.     Referenced Documents
       2.1    ASTM Standards:
              D3606 Test method for Benzene and Toluene in Finished Motor and Aviation
              Gasoline by Gas Chromatography2
              D4052 Test Method for Density and Relative  Density of Liquids by Digital
              Density Meter2
             D4057 Practice for Manual Sampling of Petroleum and Petroleum Products2
             D4307 Practice for Preparation of Liquid Blends for Use as Analytical
             Standards2
             D4420 Test Method for Aromatics in Finished Gasolines by Gas
             Chromatography2
             D4626 Practice for Calculation of Gas Chromatographic Response Factors2
             D1298 Test Method for Density, Relative Density (Specific Gravity), or API
             Gravity of Crude Petroleum and Liquid Petroleum Products by Hydrometer
             Method5

3.     Terminology
      3.1    Descriptions of Terms Specific to Tfiis Standard:
             3.1.1   low volume connector--^ special union for connecting two lengths of
                   tubing 1.6 mm inside diameter and smaller.  Sometimes this is referred to
                   as  zero dead volume union.
             3.1.2   A/7B£--methyl tertiary-butyl ether.
             3.1.3   ETBE-ethyl tertiary-butyl ether
             3.1.4   L4Af£'--tertiarv-amvl methyl ether
             3.1.5   DIPE -diisopropyl ether
             3.1.6   temflrv-Qmv/fl/co/K?/--tertiary-pentanol
             2.1.7   Oxygenate—any oxygen-containing organic compound which can be used
                   as a fuel or  fuel supplement, for example, various alcohols and ethers.
             3.1.8   split ratio~in capillary gas Chromatography, the ratio of the total flow of
                   carrier gas to the sample inlet versus the flow of the carrier gas to the
                   capillary column, expressed by

                       split ratio = (S + C)/C                                       [1]

                   where S is the flow rate at the splitter vent,
                   and C is the flow rate at the column  outlet.

                                        66

-------
              3.1.9  rC£/>--l,2,3-tris-2-cyanoethoxypropane-a'gas chromatographic liquid
                    phase.
              3.1.10 WCOT~a type of capillary gas chromatographic column prepared by
                    coating the inside of the capillary with a thin film of stationary phase.

 4.    Summary of Test Method
       4.1    An appropriate internal standard such as 1,2-dimethoxyethane (ethylene glycol
              dimethyl ether) is added to the sample which is then introduced into a gas
           •-..   chromatograph equipped with two columns and a column switching valve.  The
              sample first passes onto a polar TCEP column which elutes lighter hydrocarbons
              to vent and retains the oxygenated and heavier hydrocarbons. After
              methylcyclopentane, but before DIPE and MTBE elute from the polar column.
              the valve is switched to backflush the oxygenates onto a WCOT non-polar
          " .  column. The alcohols and ethers elute from the non-polar column in boiling
              point order, before elution of any major hydrocarbon constituents. After
              benzene and TAME elute from the non-polar column, the column switching
             valve is switched back to its original position to backflush the heavy
             hydrocarbons. The eluted components are detected by a flame ionization or
             thermal conductivity detector. The detector response, proportional to the
             component concentration, is recorded; the peak areas are measured; and the
             concentration of each component is calculated with reference to the internal.
             standard.

5.     Significance and Use
       5.1    Ethers, alcohols and other oxygenates can be added to gasoline to increase
             octane number and to reduce emissions.  Type and concentration of various
             oxygenates are specified and regulated to ensure acceptable commercial
             gasoline quality. Drivability, vapor pressure, phase separation, exhaust and
             evaporative emissions are some of the concerns associated with oxygenated
             fuels.

       5.2    This test method is applicable to both quality control in the production of
             gasoline and for the determination of deliberate or extraneous oxygenate
             additions or contamination.

6.      Apparatus
       6.1    Chromatograph:
             6.1.1  While any gas chromatographic system, which is capable of adequately
                   resolving the individual ethers and alcohols that are presented in Table 2.
                   can be used for these analyses, a gas chromatographic instrument which
                   can be operated at the conditions  given in Table 1, and having a column
                   switching and backflushing system equivalent to Fig. 1 has been found
                   acceptable. Carrier gas flow controllers shall be capable of precise
                   control where the required flow rates are low (Table 1).  Pressure control

-------
              devices and gauges shall be capable.of precise control for the typical
              pressures required.
        6.1.2  Detector-A. thermal conductivity detector or flame ionization detector,
              can be used. The system shall have sufficient sensitivity and stability to
              obtain a recorder deflection of at least 2 mm at a signal-to-noise ratio of
              at least 5 to 1 for 0.005 volume% concentration of an oxygenate.
        6.1.3  Switching and Backflushing Valve-A. valve, to be located within the gas
              chromatographic column oven, capable of performing the functions
 :             described in Section 11 and illustrated in Fig. 1. The valve shall be of low
              volume design and not contribute significantJy to chromatographic
              deterioration.
              6.1.3.1     Valco Model No. A 4C10WP, 1.6 mm (1/16 in.) fittings. This
                        particular valve was used in the majority of the analyses used
                        for the development of Section 15.
              6.1.3.2    . Valco Model No. C10W, 0.8 mm (1/32 in.) fittings. This valve
                        is recommended for use with columns of 0.32 mm inside
                        diameter and smaller.
              6.1.3.3     Some gas chromatographs are equipped with an auxiliary
                        oven which can be used to contain the valve and polar
                        column. In such a configuration, the nonpolar column is
                        located in the main oven and the temperature can be adjusted
                        for optimum oxygenates resolution.
       6.1.4   An automatic valve switching device must be used to ensure repeatable
              switching times.  Such a device should be synchronized with  injection and
              data collection times.
       6.1.5  Injection System-The, chromatograph should be equipped with a
             splitting-type inlet device if capillary columns or flame ionization
             detection are used. Split injection is necessary to maintain the actual
             chromatographed sample size within the limits of column and detector
             optimum efficiency and linearity.
             6.1.5.1     Some gas chromatographs are equipped with on-column
                        injectors and autosamplers which can inject small samples
                        sizes. Such injection systems can be used provided that
                        sample size is within the limit of the column and detectors
                        optimum efficiency and linearity.
             6.1.5.2     Microlitre syringes, automatic syringe injectors, and liquid
                        sampling valves have been used successfully for introducing
                        representative samples into the gas chromatographic inlet.

6.2    Data Presentation or Calculation, or Both:
      6.2.1  Recorder-A recording potentiometer or equivalent with a full-scale
             deflection of 5 mV or less can be used to monitor detector signal.  Full-
             scale response time should be 1 s or less with sufficient sensitivity  and
             stability to meet the requirements of 6.1.2.
                                   69

-------
              6.2.2  Integrator or Computer-Means shall be provided for determining the
                    detector response. Peak heights or areas can be measured by computer,
                    electronic integration or manual techniques.

       6.3    Columns, two as follows:
              6.3.1  Polar column-This column performs a preseparation of the oxygenates
                    from volatile hydrocarbons in the same boiling point range. The
                    oxygenates and remaining hydrocarbons are backfJushed onto the non-
                    polar column in 6.3.2. Any column with equivalent or better
                    chromatographic efficiency and selectivity to that described in 6.3.1.1 can
                    be used. The column shall perform at the same temperature as required
                    for the column in 6.3.2. except if located in a separate auxiliary oven as in
                    6.1.3.3.
                    6.3.1.1  TCEP Micro-Packed Column4,560 mm (22 in.) by 1.6 mm (1/16
                           in.) outside diameter by 0.38 mm (0.015 in.) inside diameter
                           stainless steel tube packed with 0.14 to 0.15 g of 20%
                           (mass/mass) TCEP on 80/100 mesh Chromosorb P(AW). This
                           column was used in the cooperative study to provide the
                           precision and bias data referred to in Section 15.
             6.3.2  Non-polar (Analytical) Column—Any column with equivalent or better
                   chromatographic efficiency and selectivity to that described in 6.3.2.1 and
                   illustrated in Fig. 2 can be used.
                   6.3.2.1  WCOTMethyl Silicone Column, 30m (1181 in.) long by 0.53 mm
                           (0.021 in.) inside diameter fused silica WCOT column with a
                           2,6fj.m film thickness of cross-linked methyl siloxane. This
                           column was used in the cooperative study to provide the
                           precision and bias data referred to in Section 15.

7.     Reagents and Materials
      7.1    Carrier Gas-Carrier gas appropriate to the type of detector used. Helium has
             been used successfully. The minimum purity of the carrier gas used must be
             99.95 mol%.

      7.2    Standards for Calibration and Identification—Standards of all components to be
             analyzed and the internal standard are required for establishing identification by
             retention time as well as calibration for quantitative measurements. These
             materials shall be of known purity and free of the other components to be
             analyzed.
      NOTE 1: Warning—These materials are flammable and can be harmful or fatal if
      ingested or inhaled.

      7.3   Preparation of Calibration Blends-Far best results, these components must be
            added to a stock gasoline, a hydrocarbon blend or petroleum naphtha, which is
            free of pxgenates (Warning-See Note 2).  Refer to Practice D 4307 for
            preparation of liquid blends.  The preparation of several different blends, at
                                            70

-------
             different concentration levels covering the concentration range of interest is
             necessary to verify system linearity. For best precision, a multipoint calibration
             procedure must be used. To do so, plot ratios of the areas of oxygenates to that
             of the internal standard (y-axis) against the ratios of the concentrations of the
             oxygenates to that of the internal standard (x:axis). These will be used to
             establish the linearity of the component response.
       NOTE 2: Warning-Extremely flammable. Vapors harmful if inhaled.

       7.4    Methylene Chloride-Used for column prepa; ition. Reagent grade free of non-
             volatile residue.
       NOTE 3: Warning-Harmful if inhaled. High concentrations can cause
             unconsciousness or death.

8.     Preparation of Column Packings
       8,1    TCEP Column Packing:
             8.1.1  Any satisfactory method, used in the practice of the art that will produce
                   a column capable of retaining the Cl to C4 alcohols and MTBE, ETBE,
                   DIPE and TAME from components of the same boiling point range in a
                   gasoline sample. The following procedure has been used successfully.
             8.1.2  Completely dissolve 10 g of TCEP in 100 mL of methylene chloride.
                   Next add 40 g of 80/100 mesh Chromosorb P(A W)6 to the TCEP
                   solution.  Quickly transfer this mixture to a drying dish, in a fume hood,
                   without scraping any of the residual packing from the sides of the
                   container. Constantly, but gently, stir  the packing until all of the solvent
                   has evaporated. This column packing can be used immediately to
                   prepare the TCEP column.

9.    Preparation of Micro-packed TCEP Column
      9.1     Wash a straight 560 mm length of 1.6 mm outside diameter (0.38 mm inside
             diameter) stainless steel tubing with methanol and dry with compressed
             nitrogen.

      9.2     Insert 6 to 12 strands of silvered wire, a small mesh screen or stainless steel frit
             inside one end of the tube.  Slowly add 0.14 to 0.15 g of packing material to the
             column and gently vibrate to settle  the packing inside the column. When strands
             of wire are used to retain the packing material inside the column,, leave 6.0 mm
             (0.25 in.) of space at the top of the  column.

      9.3     Coluai*. Conditioning-Both the TCEP and WCOT columns are to be briefly
             conditioned before use. Connect the columns to the valve  (see 11.1) in the
             chromatographic oven.  Adjust the carrier gas flows as in 11.3 and place the
             valve in the RESET position. -After several minutes, increase the column oven
             temperature to 120°C and maintain these conditions for 5 to 10 min.  Cool the
             columns below 60°C before shutting off the carrier flow.

                                         71

-------
10.   Sampling                                                   .
      10.1'   Every effort should be made to insure that the sample is representative of the
             fuel source from which it is taken. Follow the recommendations of Practice
             D4057 or its equivalent when obtaining samples from bulk storage or pipelines.

      10.2   Upon receiyt in the laboratory, chill the sample in its original container to 32 to
             40°F before any subsampling is performed.

      10.3   If necessary, transfer the chilled sample to a vapor tight container and store at
             32 to  40°F until needed for analysis.

11.   Preparation  of Apparatus and Establishment of Conditions
      11.1  Assembly-Connect the WCOT column to the valve system using low volume
             connectors and narrow bore tubing. It is important to minimize the volume of
             the chromatographic system that comes in contact with the sample, otherwise
             peak broadening will occur.

      11.2  Adjust the operating conditions to those listed in Table 1, but do not turn on the
            detector circuits. Check the system for leaks before proceeding further.
             11.2.1 If different polar and nonpolar columns are used, and/or capillary
                  columns of smaller ID are used, it can be necessary to use different
                  optimum flows and temperatures.

      11.3   Flow Rate Adjustment:
            11.3.1 Attach a flow measuring device to the column vent with the valve in the
                  RESET position and adjust the pressure to the injection port to give 5.0
                  mL/min flow (14 psig). Soap bubble flow meters are suitable.
            11.3.2 Attach a flow measuring device to the split injector vent and adjust the
                  flow from the split vent using the A flow controller to give a flow of 70
                  mL/min. Recheck the column vent flow set in 11.3.1 and adjust if
                  necessary.
            11.3.3  Switch the valve to the BACKFLUSH position and adjust the variable
                  restrictor to give the same column vent flow set in 11.3.1. This is
                  necessary to minimize flow changes when the valve is switched.
            11.3.4  Switch the valve to the inject position RESET and adjust the B flow
                  controller to give a flow of 3.0 to 3.2 mLVmin at the detector exit. When
                  required for the particular instrumentation used, add makeup flow or
                  TCD switching flow to give a total of 21 mL/min at the detector exit.

      11.4   When a thermal conductivity detector is used, turn on the filament current and
            allow the detector to equilibrate.  When a flame ionization detector is used, set
            the hydrogen and air flows and ignite the flame.

      11.5   Determine the Time to Blackflush~The time to backflush will vary slightly for
            each column system and must be determined experimentally as follows. The

-------
             start time of the integrator and valve timer must be synchronized with the
             injection to accurately reproduce the backflush time.
             11.5.1  Initially assume a valve BACKFLUSH time of 0.23 min. With the valve
                   RESET, inject 3 /zL of a blend containing at least 0.5% or greater
                   oxygenates (7.3), and simultaneously begin  timing the analysis. At 0.23
                   min, rotate the valve to the BACKFLUSH position and leave it there
                   until the complete elution of TAME is realized.  Record this time as the
                   RESET time, which is the time at which the valve is returned to the
                   RESET position. When all of the remaining hydrocarbons are
                   backflushed the signal will return to a stable baseline and the system is
                   ready for another analysis.  The chromatogram should appear similar to
                   the one illustrated in Fig. 2.
                   11.5.1.1.  Ensure that the BACKFLUSH time is sufficient to
                           quantitatively transfer the higher concentrations of the ethers,
                           specifically MTBE, into the  nonpolar column.
             11.5.2  It is necessary to optimize the valve BACKFLUSH time by analyzing a
                   standard blend containing oxygenates. The correct BACKFLUSH time
                   is determined experimentally by using valve switching times between 0.2
                   and 0.3 min. When the valve is switched too soon, C5 and lighter
                   hydrocarbons are backflushed and are co-eluted in the C4 alcohol section
                   of the chromatogram. When the valve BACKFLUSH is switched too
                   late, part or all of the ether component (MTBE,  ETBE or TAME) is
                   vented resulting in an incorrect ether measurement.
                   11.5.2.1.   DIPE may require a BACKFLUSH time slightly shorter than
                            the other ethers. The system may require reoptimization if the
                            analysis of DIPE is required.
             11.5.3  To facilitate setting BACKFLUSH TIME, the column vent in Figure 1
                   can be connected to a second detector (TCD or FID) as described in
                   Test Method D4420 and used to set BACKFLUSH TIME based on the
                   oxygenates standard containing the ethers of interest.

12.    Calibration and Standardization
      12.1   Identification-Determine the retention time of each component by injecting
            small amounts either separately, or in known mixtures or by  comparing the
            relative retention times with those in Table 2.
             12.1.1  In order to ensure that miminum  interference from hydrocarbons, it is
                   strongly recommended that a fuel devoid of oxygenates be
                   chromotographed to determine the level of any hydrocarbon
                   interference.

      12.2   Preparation of Calibration Samples--Several calibration blends of the specific
            components being analyzed in an appropriate reference fuel or standard
            hydrocarbon blend should  be prepared and analyzed to establish whether the
            response is linear over the  range of concentrations under study. In the
            preparation of blends, the  total mass percent of oxygenated  components in the

                                         73

-------
              reference fuel or standard hydrocarbon blend including the internal standard,
              must not exceed 30 mass percent.

       12.3   Standardization-Trie area under each peak in the chromatogram is considered a
              quantitative measure of the corresponding compound. Measure the peak area
              of each oxygenate and of the internal standard by either manual methods or
              electronic integrator. Calculate the response of each oxygenate, relative to the
              internal standard, according to Practice D 4626.2

 13.    Procedure
       13.1   Preparation of sample—Add a quantity of internal standard to an accurately
              measured quantity of chilled sample on a gravimetric (mass) basis. The diether,
              1,2- dimethoxyethane, has been found to be an appropriate internal standard. A
              diether concentration in the 5 to 7 mass% range has been used successfully.
              This corresponds to 1.8 to 2.5 mass% added oxygen.

       13.2    Chromatographic Analysis-Introduce, a representative aliquot of the sample,
             containing internal standard, into the gas chromatograph using the same
             technique and sample size as used for the calibration analysis. An injection
             volume of 3.0 ^il with a 15:1 split ratio has been used successfully. Start
             recording and integrating devices in synchronization with sample introduction.
             Obtain a chromatogram or integrated peak report or both which displays the
             retention times and integrated area of each detected component.

       13.3   Interpretation of Chromatogram—Compare the retention times of sample
             components to those of the calibration analysis to determine the identities  of
             oxygenates present.
14.    Calculation
      14.1  Moss Concentration of Oxygenates- After identifying the various oxygenates
            measure the area of each oxygenate peak and that of the internal standard.
            Calculate the mass% of each oxygenate according to equation 2 as follows:

                        Ws x AJ x RJ x 100                                  -
      where:     ^^
      Xi     =     mass% of each oxygenate being determined,

      Ws    =     mass of internal standard (1,2-dimethoxyethane) added,

      Wg    =     mass of gasoline sample taken,
                                         74

-------
                                        10


 Ai     =     peak area of the oxygenate to be deteimined.

 As     =     peak area of the internal standard (1,2-dimetnoxyethane), and

 Ri     =     mass relative response factor of each component (relative to the internal
              standard).

 14.2   Report the mass% of each oxygenate to the nearest 0.01 mass %.

 14.3  Mass % oxygen--To determine the oxygen content of the fuel, convert and sum
        the oxygen contents of all oxygenated components determined in 14.1 according
       to equation 3 as follows:
                      Xjxl6.0xNj
                                                                            [3]
             or
                              N!    X2xl6.0xN2
                                  +   "  M2    ~ + .....................              [

where:
Xj     =     mass% of each oxygenate

^tot   ~     tota' mass% oxygen in the fuel,

Mj     =     molecular weight of the oxygenate as given in table 2, and

16.0   =     atomic weight of oxygen,

NJ     =     number of oxygen atoms in the oxygenate molecule

14.4   Report the total mass% of oxygen in the fuel to the nearest 0.01 mass %.

14.5   Volumetric Concentration of Oxygenates~lf the volumetric concentration of each
       oxygenate is desired, calculate the volumetric concentrations according to
       equation 5 as follows:
                                                                            [5]
where:
Vj     =     volume % of each oxygenate to be determined,

-------
                                              11
       Wj    =     mass% of each oxygenate as calculated from equation 1,

       Dj     =     relative density at 60°F (15.56°C) of the individual oxygenate as found in
                    table 2,

       Df    =     relative density of the fuel under study as determined by Test Methods
                    D-1298 or D-4052

       14.6   Report the volume% of each oxygenate tc the nearest C;01 volume %.
 15.    Precision and Bi
       15.1   Precision—The, precision of this test method as determined by a statistical
             examination of interlaboratory test results is as follows:
             15.1.1  Repeatability-Tile difference between successive results obtained by the
                    same operator with the same apparatus under constant operating
                    conditions on identical test materials would, in the long run, in the normal
                    and the correct operation of the test method exceed the following values
                    only in one case in twenty (see Table 3).
Supporting data are/will be available from ASTM Headquarters. Request RR:D02-xxxx.
                                                  76

-------
                                         12
                                     TAB LEI
                  CHROMATOGRAPHIC OPERATION CONDITIONS
           ' Temperatures
     Column Oven
     Injector, °C
     Detector-TCD, °C
            -FID, °C
    • Valve °C  '
Flows, ml/min
Carrier Gas: Helium
60
200
200
250
60
to injector
Column
Auxiliary
Makeup

75
5
3
18

                Sample size,
                Split ratio
                Backflush. min
                Valve reset time
                Total Analysis time
                1.0-3.0
                  15:1
              .  0.2-0.3
              8-10 min
             18-20 min
A) Sample size must be adjusted so that alcohols in the range of 0.1 to 12.0 mass % and
   ethers in the range of 0.1 to 20.0 mass % are eluted from the column and measured
   linearly at the detector.  A sample size of 1.0 /^L has been introduced in most cases.
                                       77

-------
                                         13
                                     TABLE 2

                        PERTINENT PHYSICAL CONSTANTS AND
                          RETENTION CHARACTERISTICS FOR
                  TCEP/WCOT COLUMN SET CONDITIONS AS IN TABLE 1
       Component

Water
Methanol
Ethanol
Isopropanol
tert-Butanol
n-Propanol
MTBE
sec-Butanol
DIPE
Isobutanol
ETBE
tert-Pentanol
1,2-Dimethoxyethane (DME)
ri-Butanol
Benzene
TAME

Retention
Time, Min.
2.90
3.15
3.48
3.83
4.15
4.56
5.04
5.36
5.76
6.00
6.20
6.43
6.80
7.04
7.41
8.17

Relative
(MTBE - 1.
0.58
0.63
0.69
0.76
. 0.82
0.90
1.00
1.06
1.14
1.19
1.23
1.28
1.35
1.40
1.47
1.62

Retention Time
00) ( DME =1.00)
0.43
0.46
0.51
0.56
0.61
0.67
0.74
0.79
0.85
0.88
0.91
0.95
1.00
1.04
1.09
1.20

Molecular
Weight
18.0
32.0
46.1
60.1
74.1
60.1
88.2
74.1
102.2
74.1
102.2
88.1
90.1
74.1
78.1
102.2
Relative
Density' at
1 5.56/1 5.56T
1.000
0.7963
0.7939
0.7899
0.7922
0.8080
0.7460
0.8114
0.7300
0.8058
0.7452
0.8170.
0.8720
0.8137
0.8830
0.7758
                                          78

-------
                                                             14
                                                              TABLES
                                   PRECISION INTERVAL AS DETERMINED FROM
                                               COOPERATIVE STUDY DATA
 Component  MEOH
         ElOH    iPA
                 tBA
                         nPA
                                 Repeaubiljr
                          MTBE    sBA
                                      iBA    ETBE
                                                                      tAA
                                                      nBA
                                                     TAME
                                                     Total
                                                     Oxygen
Wt.%
0.20
0.50
1.00
2.00
3.00
4.00
5.00-
6.00
10.00
12.00
14.00
16.00
20.00

0.03
0.04
0.06
0.09
0.11
0.12
0.14
0.15
0.19
0.21




0.01 0.01
0.02 0.02
0.03 0.02
0.04 0.02
0.05
0.06
0.07
0.08
0.10
0.11




0.01
0.02
0.03
0.03
0.07
0.08
0.09
0.09
0.15
0.18




0.002
0.003
0.004
0.01









                                                          0.01
                                                          0.01
                                                          0.02
                                                          0.03
                                                          0.04
                                                          0.05
                                                          0.06
                                                          0.07
                                                          0.09
                                                          0.10
                                                          0.11
                                                          0.13
                                                          0.14
                                                     Q.W2
                                                     0.004
                                                     0.01
                                                     0.01
                                                      0.02
                                                      0.04
                                                      0.07
                                                      0.11
                                                      0.02
                                                      0.03
                                                      0.05
                                                      0.08
                                                      0.11
                                                      0.14
                                                      0.17
                                                      0.19
                                                      0.28
                                                      0.32
                                                      0.36
                                                      0.40
                                                      0.47
                                                      0.03
                                                      0.05
                                                      0.07
                                                      0.11
                                                      0.05
                                                      0.06
                                                      0.07
                                                      0.08
                                                      0.005
                                                      0.01
                                                      0.01
                                                      0.01
                                                      0.02
                                                      0.02
                                                      0.02
                                                      o.o:
                                                      0.03
                                                      0.03
                                                      0.03
                                                      0.04
                                                      0.04
                                                      o.o:
                                                      0.04
                                                      0.05
                                                      0.07
                                                      0.03
Component   MEOH
         ElOH    IPA
                          tBA
                         nPA
                                       Reprod liability
                                 MTBE    sBA      iBA
                                                     ETBE
                                                                                             tAA
                                                              nBA
                                                              TAME
                                                              Total
                                                              Oxvgen
    Wt.%
   0.20
   0.50
   1.00
   2.00
  . 3.00
   4.00
   5.00
   6.00
  10.00
  12.00
  14.00
  16.00
  20.00
0.17
0.29
0.44
0.67
0.86
1.02
1.17
1.31
1.79
2.00
0.10
0.20
035
0.61
0.83
1.05
1.25
1.44
2.16
2.49
0.07
0.17
0.32
0.59
0.04
0.09
0.18
0.35
0.51
0.67
0.83
0.99
1.60
1.91
Z21
Z51
3.10
0.02
0.03
0.05
0.07
0.05
0.12
0.23
0.45
0.66
0.87
1.08
1.29
2.10
2JO
2.90
3.29
4.08
0.11
0.12
0.13
0.14
0.01
0.04
0.08
0.18
0.19
034
0.54
0.85
1.12
1.35
1.56
1.76
2.47
2.78
3.08
337
3.90
0.06
0.15
0.29
0.57
0.13
0.22
0.32
0.47
0.20
034
0.50
0.74
0.94
1.10
1.25
139 .
1.86
2.06
125
2.43
2.76
0.26
0.45
0.61
0.77
0.91
                                                        79

-------
                                     15
      REPEATABILITY ESTIMATES FOR OXYGENATES IN GASOLINE

            Component                                Repeatability
            Methanol (MeOH)                         0.06 (X0-51)
            Ethanol (EtOH)                            0.03 (X0-52)
            Isopropanol (iPA)                          0.02 (X^-25)
            tert-Butanol (tBA)                          0.03 (X0-71)
            n-Propanol (nPA)                     '.    0.004 (X0-47)
            MTBE                                    0.02 (X°-66)
            sec-Butanol (sBA)                          0.005 (X^-53)
            Isobutanol (iBA)                      .      0.07 (X0-67)
            ETBE                                    0.05(XO-75)
            tert-Pentanol (tAA)                         0.07 (X0-64)
            n-Butahol (nBA)                            0.07 (X^-^)
            TAME                                    0.01(XO-45)
            Total Oxygen                               0.02 (X0-88)
where X is the mean mass percent of the component.

15.1.2 Reproducibility-The difference between two single and independent results
      obtained by different operators working in different laboratories on identical
      material would, in the long run, exceed the following values only in one case in
      twenty (see Table 3).

-------
                                           16
           REPRODUCIBILITY ESTIMATES IN OXYGENATES IN GASOLINES

               Component                                Reproducibility
              ' Methanol (MeOH)                          0.44 (X0-61)
               Ethanol (EtOH)                            0.35 (X0-79)
               Isopropanol (iPA)                           0.32 (X^-88)
               tert-Butanol (tBA)                          0.18 (X0-95)
               n-Propanol (nPA)                           0.05 (X^-56)
               MTBE                                    0,23 (XO-96)
               sec-Butanol (sBA)                           0.13 (X0-09)
               Isobutanol (iBA)                            0.08 (X^-20)
               ETBE                                     .0.54 (X0-66)
               tert-Pentanol (tAA)                          0.29 (X0-97)
               n-Butanoi (nBA)                            0.32 (X0-55)
               TAME                                    0.50 (X0.57)
               Total Oxygen                               0.26 (X0-78)

          where X is the mean mass percent of the component.

      15.2   Bias~The National Institutes of Standards and Technology (NIST) provides
            selected alcohols in reference fuels. As an example the following standard
            reference materials (SRM) in reference fuels are available as described in the
            NIST Standard Reference Catalog.^
                                            Nominal Concentration, Mass % of
     SRM    Ty£e                         MeOH      EtOH    MeOH + tBuOH
     1829    Alcohols in Reference Fuel       0.335       11.39     "  10.33 + 6.63
     1837    Methanol and tert-butanol                              10.33 + 6:63
     1838    Ethanol                             .      11.39
     1839    Methanol                      07335

16.    Keywords
      MTBE (Methyl tert-butylether)                               •
      ETBE (Ethyl tert-butylether)
      TAME (Tert-amylmethylether)
      DIPE (Disopropylether)
      Alcohols
      Ethers
      Oxygenates
      Gasoline
      Gas Chromatography

-------
                                             17
                                    FOOTNOTES

 1.  '  This test method is under the jurisdiction of ASTM Committee D-2 on Petroleum
       Products and Lubricants and is the direct responsibility of Subcommittee D02.04 on
       Hydrocarbon Analysis.

       Current edition approved October 27, 1989. Published December 1989. Orginally
       published as D 4815-88. Last previous edition D 4815-88.

 2.     Annual Book of ASTM Standards, Vol 05.03.

 3.     Annual Book of ASTM Standards, Vol 14.01.

 4.     Available from Hewlett Packard Company, Avondale, PA.

 5.     Annual Book of ASTM Standards, Vol 05.01

 6.     Available from Supelco Inc., Bellefonte, PA

 7.     Supporting data are/will be available from ASTM Headquarters, Request RR: D)2-
       xxxx.

 8.      NIST Special Publication 260; NIST Standard Reference Materials 1990-1991

 First Draft, Sub.D02.04L, 12/19/91

 Second Draft, Sub D02.04L, 1/31/92

 Third Draft, Sub.D02.04L, 2/28/92

 Fourth Draft, Sub.D02.04L, 7/7/92

 Fifth Draft, Sub. D02.04, 8/4/92

Sixth Draft, Sub D02.04, 8/20/92 '                                          .
                                        82

-------
    V~ wt, w
 u
 ki
 l_
 id
 C
                                       POLAR i 7CZ?;

  XQN-FOLAR
'•fCCT CCL'JIW
                                         NJECTCR
                       X°N..
                       .  A
          Va i ve  :n  H£SZT  =cs; : :on
                  AD.'USTAELE
                  ZTQT'y r (—-^p
                  • tm**Mt * > * i W t fc-f * *
CCL.LIUN VENT
                              FOLXR
                                CCLL1JN
  U*CCT CCL'JWN
                           IN
                                       SPLI
                                    SPLIT
                                     1NJECTCH
                      A          4
                      i	 
-------
       c  -
z
c
*H*
m
              f
      -  1
          I    I
     01
     o
                                                    sec-5utanol
                                                    DI?E
                                                   	-. Isobutanol
                                                        :=»  cert-Pencasox
                                                    r.-5utanol
                                                    cer.zene
                           Valve  Reset
    o  J
    c   '

-------
                   Appendix C - Test Procedure
       Test for the Determination of Oxygenates in Gasoline

 1.  Scope  and  application.
 1.1  The following single-column, direct-injection gas
 chromatographic procedure  is described in detail to completely
 define a single technique  for quantifying the oxygenate content
 of gasoline.   Other procedures with similar capabilities are
 allowed provided they comply with the quality control
 requirements of section 8  below.
 1.2  This  method covers the quantitative determination of the
 oxygenate  content of gasoline through the use of an oxygenate
 flame ionization detector  (OFID).  It is applicable to individual
 organic oxygenated compounds (up to 20 mass percent each) in
 gasoline having a final boiling point not greater than 220 °C.
 Samples above  this level should be diluted to fall within the
 specified  range.
 1.3  Where trade names or  specific products are noted in the
 method, equivalent apparatus and chemical reagents may be used.
 Mention of trade names or  specific products is for the assistance
 of the user and does not constitute endorsement by the U.S.
 Environmental  Protection Agency.

 2.  Summary of method.
 2.1  A sample  of gasoline  is spiked to introduce an internal.
 standard, mixed,  and injected into a gas chromatograph (GC)
 equipped with  an OFID.  After chromatographic resolution the
 sample components enter a  cracker reactor in which they are
 stoichiometrically converted to carbon monoxide (in the case of
 oxygenates), elemental carbon,  and hydrogen.  The carbon monoxide
 then enters a  methanizer reactor for conversion to water and .
 methane.  Finally, the methane generated is determined by a flame
 ionization detector (FID).
 2.2  All oxygenated gasoline components (alcohols, ethers, etc.)
may be assessed by this method.
 2.3  The total mass percent of oxygen in the gasoline due to
 oxygenated components may  also be determined with this method by
 summation of all peak areas except for dissolved oxygen, water,
 and the internal standard.
 3.  Sample handling and preservation.
 3.1  Samples shall be collected and stored in containers which
will protect them from changes in the component contents of the
gasoline, such as loss of  volatile fractions of the gasoline by
evaporation.
 3.2  If samples have been  refrigerated they shall be brought to
 room temperature prior to  analysis.
 3.3  Gasoline  is extremely flammable and should be handled
cautiously and with adequate ventilation.  The vapors are harmful
 if inhaled and prolonged breathing of vapors should be avoided.
 Skin contact should be minimized.

 4.  Apparatus.

                                85

-------
 4.1  A GC equipped with an oxygenate flame ionization detector.
 4.2  An autosampler for the GC is highly recommended.
 4.3  A 60-m length, 0.25-mm ID,  1.0-/M film thickness, nonpolar
 capillary GC column (J&W DB-1 or equivalent)  is recommended.
 4.4  An integrator or other acceptable system to collect and
 process the GC signal.

 5.   .Reagents and Materials.
   Note:  Gasoline and many of the oxygenate additives are
 extremely flammable and may be toxic over prolonged exposure.
 Methanol is particularly hazardous.   Persons performing  this
 procedure must be familiar with the chemicals involved and all
 precautions applicable to each.
 5.1  Reagent-grade oxygenates for internal standards and for
 preparation of standard solutions.
 5.2  Supply of oxygenate-free gasoline for blank assessments and
 for preparation of standard solutions.
 5.-3  Calibration standard solutions  containing known quantities
 of  suspected oxygenates in gasoline.
 5.4  Calibration check standard solutions prepared  in the same
 manner as the calibration standards.
 5.5  Reference standard solutions containing known  quantities of
.suspected oxygenates in gasoline.
 5.6  Glass standard and test sample  containers (between  5 and 100
 mL  capacity)  fitted with a self-sealing polytetrafluoroethlene
 (PTFE)  faced rubber septum crimp on  or screw down sealing cap for
 preparation of standards and samples.

 6.   Calibration.
 6.1  Calibration standards of reagent-grade or better oxygenates
 (such as methanol,  absolute ethanol, methyl t-butyl  ether (MTBE),
 di-i-propyl ether (DIPE),  ethyl  t-butyl ether (ETBE),  and t-amyl
 methyl  ether (TAME))  are to be prepared gravimetrically  by
 blending  with gasoline that has been  previously determined by
 GC/OFID to be free  of oxygenates.  Newly acquired stocks of
 reagent grade oxygenates shall be analyzed for contamination by
 GC/FID  and GC/OFID  before use.
 6.2  Tare a glass sample container and its PTFE faced rubber
 septum  sealing cap.   Transfer .a  quantity of an oxygenate to the
 sample  container through the septum  and record the mass  of the
 oxygenate to the nearest 0.1 mg.  Repeat this process for any
 additional oxygenates of interest except the  internal  standard.
 Add oxygenate-free  gasoline to dilute  the oxygenates to  the
 desired concentration.   Record the mass of gasoline  added to the
 nearest 0.1 mg,  and determine and label the standard according to
 the mass percent quantities of each  oxygenate added.   These
 standards are not to exceed 20 mass  percent for any  individual
 pure component due  to potential  hydrocarbon breakthrough and/or
 loss of calibration linearity.
 6.3   Inject a quantity  of an internal  standard (such as
 2-butanol)  through  the  rubber septum and weigh the contents
 again.   Record the  difference in masses as the mass  of internal

                                86

-------
 standard to  the nearest  0.1 mg.  The mass  of  the  internal
 standard shall  amount  to between 2  and  6 percent  of  the mass  of
 the test sample (standard) .  The addition  of  an internal standard
 reduces  errors  caused  by variations in  injection  volumes.
 6 . 4  Ensure  that the prepared  standard  is  thoroughly mixed  and
 transfer approximately 2 mL of the  solution through  the septum of
 a  vial compatible with the autosampler  if  such equipment is used.
 6.5  Based on the recommended  chromatographic operating
 conditions  (section 7.1  below), determine  the retention time  of
 each oxygenate .component by analyzing dilute  aliquots either
 separately or in known mixtures.  Reference should be made  to the
 Chemical Abstracts Service  (CAS) registry  number  of  each of the
"analytes for proper identification.  Approximate  retention  times
 for selected oxygenates   under these conditions are  as follows:

      Oxygenate               CAS              Retention time
                                                minutes

      Dissolved  oxygen     7782-44-7               3.85
      Water                7732-18-5               4.28
      Methanol               67-56-1               4.71
      Ethanol                64-17-5-              5.66
      Propanone              67-64-1               6.29
      2-Propanol             67-63-0               6.63
      t-Butanol              75-65-0               7.46
      n-Propanol             71-23-8               8.60
      MTBE                 1634-04-4               9.41
      2-Butanol            15892-23-6               10.58
      DIPE                .  108-20-3               11.30
      i-Butanol              78-83-1               12.30
      ETBE                  637-92-3               12.50
      t-Pentanol             75-85-4               13.23
      n-Butanol.      .        71-36-3               14.40
      TAME                  994-05-8               15.40
      i-Pentanol   .         137-32-6               17.19

6.6   By GC/OFID  analysis, determine the peak  area of each
oxygenate and of the internal  standard.
6.7  Obtain a calibration curve by performing a least-squares fit
of the relative  area response  factors of the  oxygenate standards
to their relative mass response factors as follows:

     Rao   =   bo^+M1^)2

where:
Rao =  relative area response factor of the oxygenate, A0/Ai
      relative mass response factor of the oxygenate , M^Mi
      area of the oxygenate peak
      area of the internal standard peak
      mass of the oxygenate added to the calibration standard
      mass of internal standard added to the  calibration
                                87

-------
       standard
 bfl  =  linear regression coefficient
 b1  =  quadratic regression coefficient
 6.8  Before proceeding with the analysis  of  samples,  the
 least-squares regression should be evaluated for  goodness-of-fit.

 7.   Procedure.
 7.1  GC operating conditions:
 7.1.1  Oxygenate-free helium carrier gas:  1.1 mL/min  (2 bar),
 22.7 cm/sec at ,115 °C
 7.1.2  Carrier gas split ratio:  1:100
 7.1.3:  Zero-air FID fuel:  370  mL/min (2 bar)
 7.1.4  Oxygenate-free hydrogen FID fuel:  15  mL/min  (2 bar)
 7.1.5  Injector temperature:  250 °C
 7.1.6  Injection volume:   0.5  A*L
 7.1.7  Cracker reactor temperature:  sufficiently  high to ensure
 reduction  of all hydrocarbons  to the elemental  states  (i.e.,
 C,H2x->C+H2, etc.)
 7.1.8  FID temperature:  400 °C
 7.1.9  Oven temperature program:  40  °C for 6 min,  followed by a
 temperature increase of 5  °C/min to 50 °c,  hold  at 50  °C for 5
 min,  followed by a temperature increase of 25 °C/min to 175 °C,
 and hold at 175  °C for 2 min.
 7.2   Prior to analysis of  any  samples, inject a sample of
 oxygenate-free gasoline into the GC  to test  for hydrocarbon
 breakthrough overloading the cracker reactor.   If breakthrough
 occurs, the OFID is  not operating effectively and must be
 corrected  before samples can be  analyzed.
 7.3   Prepare gasoline  test samples for analysis as  follows:
 7.3.1  Tare a glass  sample container and  its  PTFE faced rubber
 septum  sealing cap.  Transfer  a  quantity of the gasoline sample
 to  the  sample container through  the  septum and record the mass of
 the  transferred  sample to  the  nearest 0.1  mg.
 7.3.2   Inject a  quantity of the  same  internal standard (such as
 2-butanol)  used  in generating  the  standards through the rubber
 septum  and  weigh the contents  again.  Record  the difference in
masses  as  the mass of  internal standard to the nearest 0.1 mg.
The mass of  the  internal standard  shall amount to between 2 and 6
percent of  the mass of  the test  sample (standard).  The addition
of an internal standard reduces  errors caused by variations in
 injection volumes.
7.3.3   Ensure that this test sample  (gasoline plus  internal
standard)  is  thoroughly mixed  and  transfer approximately 2 mL of
the solution  through the septum  of a vial  compatible with the
autosampler  if such equipment  is used.
7.4  After  GC/OFID analysis, identify the  oxygenates in the
sample  based  on  retention  times, determine the peak area of each
oxygenate and of the internal  standard, and calculate the
relative area response  factor  for  each oxygenate.
7.5  Monitor  the peak  area of  the  internal standard.  A larger
than expected peak area for the  internal standard when analyzing


                                88

-------
 a test sample may indicate that this oxygenate is present in the
 original sample.  Prepare a new aliquot of the sample' without
 addition of the oxygenate internal standard.  If the presence of
 the oxygenate previously used as the internal standard can be
 detected, then either  (1) the concentration of this oxygenate
 must be assessed by the method of standard additions or (2)  an
 alternative internal standard, based on an oxygenate that is not
 present in the original sample, should be utilized with new
 calibration curves.
 7.6  Calculate 'the relative mass response factor (R^) for each
 oxygenate based on the relative area response factor (Rao) and
 the calibration equation of section 6.7 above.
 7 . 7.  Calculate the mass percent of the oxygenate in the test
 sample according to the following equation:
 M0%    =   -----------
•  . * -       • M.
 where:
 M0%  =  mass percent of the oxygenate in the test  sample
 Ms   =  mass of sample to which internal standard  is added
 7.8  If the mass percent exceeds the calibrated range,
 gravimetrically dilute a portion of the original  sample to a
 concentration within the calibration range and analyze this
 sample starting with section 7.3 above.
 7.9  Report the total mass percent oxygen as follows:
 7.9.1  Subtract the peak areas due to dissolved oxygen, water,
 and the internal standard from the total summed peak areas of the
 chromatogram.
 7.9.2  Assume the total summed peak area solely due to one of the
 oxygenates that the instrument is calibrated for  and determine
 the total mass percent as that oxygenate based on section 7.7
 above.   For simplicity, choose an oxygenate having one oxygen
 atom per molecule.
 7.9.3  Multiply this concentration by the molar mass of oxygen
 and divide by the molar mass of the chosen oxygenate to determine
 the  mass percent oxygen in the sample.   For example,  if the total
 peak area is based on MTBE,  multiply by 16.00 (the molar mass of
 atomic oxygen)  and divide by 88.15 (the molar mass of MTBE) .
 7.10   Sufficient sample should be retained to permit reanalysis.

 8.   Quality control of precision and accuracy.
 8.1   The laboratory shall routinely monitor the repeatability
 (precision)  of its analyses.  The recommendations are:
 8.1.1  The preparation and analysis of laboratory duplicates at a
 rate  of one per analysis batch or at least one per ten samples,
 whichever is more frequent.
 8.1.2.  Laboratory duplicates should be carried through all sample
 preparat4.i*i steps independently.
 8.1.3  The range (R)  for duplicate samples should be less than
 the  following limits:


                                 89

-------
     Oxygenate
                  Concentration
                  mass percent
Upper Limit for Range
mass percent
Methanol
Methanol
Ethanol
MTBE
DIPE
ETBE
TAME
0.
1.
1.
0.
0.
1.
1.
27
07
01
25
98
00
04
to
to
to
to
to
to
to
1
12
12
15
17
18
18
.07
.73
.70
.00
.70
.04
.59
0.
0.
0.
0.
0.
0.
0.
010+0.
053C
053C
069+0.
048C
074C
060C
043C


029C



where:
      =  
-------
.8.1.7   The  relationship of standard  deviation and concentration
for raethanol  between 0.27  and 1.07 mass  percent was very linear
and is  described  by the equation:

standard deviation = 0.0118*0+0.0027
8.1.8   Based  on these relationships, repeatability for the
selected oxygenates at 2.0 and 2.7 mass  percent oxygen were
determined  to be  as follows:

OxygenateConcentration         ~~~Repeatability
            mass percent   mass percent   volume percent    mass percent
 ._           oxygen         oxygenate     oxygenate
Methanol
Ethanol
MTBE
DIPE
ETBE
TAME
Methanol
Ethanol
MTBE
DIPE
ETBE
TAME
2.0
2.0
2.0
2.0
2.0
2.0
2.7
2.7
2.7
2.7
2.7
2.7
4.00
5.75
11.00
12.77
12.77
12.77
5.40
7.76
14.88
17.24
17.24
17.24
3.75
v 5.41
11.00
13.00
12.74
12.33
5.07
7.31
14.88
17.53
17.20
16.68
0.11
0.16
0.21
0.32
0.50
0.41
0.15
0.21
0.26
0.43
0.67
0.55
where repeatability  is defined as the half width of the  95
percent confidence interval  (i.e., 1.96 standard deviations)  for
a single analysis at the  stated concentration.
8.2  The laboratory  shall routinely monitor the accuracy of  its
analyses.  The recommendations are:
8.2.1  Calibration check  standards and calibration standards  may
be prepared from the same oxygenate stocks and by the same
analyst.  However, calibration check standards and calibration
standards must be prepared from separate batches of the  final
diluted standards.   For the  specification limits listed  in
section 8.2.3, the concentration of the check standards  should be
in the range given in section 8.1.3.
8.2.2  Calibration check  standards should be analyzed at a rate
of one per analysis batch or at least one per ten samples,
whichever is more frequent.
8.2.3  If the measured concentration of a calibration check
standard is outside the range of 100.0±6.0 percent of the
theoretical concentration for a selected oxygenate of 1.0 mass .
percent or above, the sources of error in the analysis should be
determined, corrected, and all analyses subsequent to and
including the last standard analysis confirmed to be within the
compliance specifications must be repeated.  The specification
limits for the accuracy of calibration check standards analyses
are minimum performance requirements.  The performance of
individual laboratories may indeed be better than these  minimum
requirements.  For this reason it is recommended that control
charts be utilized to monitor the variability of measurements in
order to optimally detect abnormal situations and ensure a stable


                                91

-------
measurement process.
8.2.4  Independent reference standards should be purchased or
prepared  from materials that are independent of the calibration
standards and calibration check standards, and must not be
prepared by the same analyst.  For the specification limits
listed in section 8.2.6, the concentration of the reference
standards should be in the range given in section 8.1.3.
8.2.5  Independent reference standards should be analyzed at a
rate of one per analysis batch or at least one per 100 samples,
whichever is mpre frequent.
8.2.6  If the measured concentration of an independent reference
standard  is outside the range of 100.0±10.0 ^percent of the
theoretical concentration for a selected oxygenate of 1.0 mass
percent or above, the sources of error in the analysis should be
determined, corrected, and all analyses subsequent to and
including the last independent reference standard analysis
confirmed to be within the compliance specifications in that
batch must be repeated.  The specification limits for the
accuracy of independent reference standards analyses are minimum
performance requirements.  The performance of individual
laboratories may indeed be better than these .minimum
requirements.  For this reason it is recommended that control
charts be utilized to monitor the variability of measurements in
order to optimally detect abnormal situations and ensure a stable
measurement process.
                                92

-------
               Appendix D - Illustrative CPA Report

 [Address to the  CAR's  management  and  state  regulatory  agency]

 I/We  have applied the  procedures  enumerated below  to the
 accompanying [title  of the  CAR's  oxygenated fuels  report(s)  filed
 with  the state]  for  the [control  area(s)] for the  [averaging
 period(s)].   These procedures,  which  were included in  [describe
 the applicable .state regulations]  and agreed to by [the CAR'S
 management],  were performed solely to assist [the  CAR  and the
 state agency]  in evaluating [name of  CAR]'s compliance with
 reporting ou oxygenated fuels distribution  under [applicable
 state regulation(s)].   This report is intended solely  for the
 information of [the  CAR's management]  and [the applicable state
 agency]  and should not be used  by others.

 [Include paragraph to  enumerate procedures  and findings.]

 These agreed-upon procedures are  substantially less in scope than
 an examination,  the  objective of  which is the expression of  an
 opinion  on  [the  CAR's  report].  Accordingly, we do not express
 such  an  opinion.   Had  I/we  performed  additional procedures or had
 we made  an  examination of such  data,  other matter might have come
 to our attention that  would have  been reported to you.

 [Signature]        [City, State]        [Date]

 Note:  This report  is for illustrative purposes only and is based
 upon  reporting Statements on Standards for Attestation
 Engagements of the American Institute of Certified Public
Accountants °1989, which were applicable at the time of issuance
 of these guidelines..  The CPA issuing such reports under the
 applicable state  regulations should revise  it,  as necessary, to
 reflect any additional guidance issued and changes subsequently
made to those reporting standards.
                                93

-------
     Appendix E - Illustrative CIA or Internal Auditor Report

 [Address  to  the  CAR'S management  and  state regulatory agency]

 I/We have applied the procedures  enumerated below to the
 accompanying [title  of  the CAR'S  oxygenated fuels report(s)  filed
 with the  state]  for  the [control  area(s)] for the [averaging
 period(s)].   These procedures, which  were included in  [describe
 the applicable state regulations] and agreed to by [the CAR's
 management]  and  [the applicable state agency], were performed
 solely to assist [the CAR management  and the state agency] in
 evaluating [name of  CAR]'s compliance with reporting on
 oxygenated fuels distribution under  [applicable state
 regulation(s)].  This report is intended for the information of
 [the CAR's management]  and [the applicable state agency] and
 accordingly  is not addressed to and should not be used by others.

 [Include  a section to enumerate procedures and finding in
 accordance with  Codification of Standards for the Professional
 Practice  of  Internal Auditing; Std. 430:  Communicating Results.
 This section should  include the following elements:

 Criteria:The standards, measures  or expectations used in making
 an evaluationand/or  verification  (what should exist).

 Condition:The factual evidence which  the internal auditor found
 in the course of the examination  (what does exist).

Cause:The  reason for the difference between the expected and
actual conditions (why  the difference exists) .

Effect:The risk  or exposure the auditee organization and/or other
encounterbecause the condition is not the same as the criteria
 (the impact  of the difference).]

 [Signature]      [Credentials]     [Address]
 [Title]                           [Date]       :
                                94

-------