I V
W
pro"^
Transfer Standards for Calibration
of Air Monitoring Analyzers for
Ozone
Technical Assistance Document
January 2023
-------
EPA-454/B-22-003
JANUARY 2023
This page intentionally left blank
-------
EPA-454/B-22-003
JANUARY 2023
Transfer Standards for Calibration of Air Monitoring Analyzers for Ozone
Technical Assistance Document
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
Disclaimer
EPA does not endorse any product or manufacturer. Any references to commercial products or
trade names in this document are factual statements regarding types of instruments currently in
use.
-------
Table of Contents
1.0 Introduction 1
1.1 The Importance ofC>3 Measurement Traceability 1
1.2 Code of Federal Regulations (CFR) Requirement 1
1.3 Summary of Photometric Technique 1
1.4 O3 Technical Assistance Document History 2
1.5 Document Organization 3
1.6 Summary of Major Changes from the Transfer Standards for Calibration of Air Monitoring Analyzers for
Ozone (2013) 3
1.7 Use of the Terms Must and Should 4
2.0 EPA's O3 Traceability Scheme 4
2.1 Description of EPA's O3 Traceability Scheme 4
2.2 Lifecycle of O3 Transfer Standards 6
2.3 Level 1 Transfer Standard Scheme 7
2.4 Defining the Terms Level 2 and Level 3 Transfer Standard 10
2.5 Defining the Terms Bench and Field Transfer Standard 10
2.6 Bench versus Field Transfer Standard Verification Frequency 10
2.7 Lapses in Verification Frequency 12
2.8 Level 2 and 3 Transfer Standard Scheme 12
2.8.1 Scheme #1 - Use of Level 3 Bench Standard and Level 2 Field Standard 13
2.8.2 Scheme #2 - Use of Level 3 Field Standard without Level 3 Bench Standard 15
2.8.3 Scheme #3 - Use of Level 2 Bench Standard only 16
2.9 Quality Documentation 17
2.10 Training 17
3.0 Types of O3 Devices 18
3.1 Devices with Photometers 18
3.2 Generator-Only devices 18
4.0 Verification and Reverification Requirements for Os Transfer Standards 18
4.1 O3 Transfer Standard Requirements and Specifications Table 18
4.2 Requirements for all Levels 20
4.2.1 Qualification Testing Requirements 20
4.2.2 Acceptance Testing Requirements 21
4.3 Level 1 Requirements 21
4.4 Level 2 and 3 Verification and Reverification Requirements 22
-------
4.4.1 Verification Requirements
4.4.2 Reverification Requirements
4.5 Level 4 Verification and Reverification Requirements
5.0 Verification and Reverification Procedure for O3 Transfer Standards
5.1 Verification Procedure
5.2 Reverification Procedure
6.0 Operational Considerations of Os Transfer Standards
6.1 How to Troubleshoot a Verification, Calibration, or Measurement Quality Check Exceeding an Action
Limit or Acceptance Criteria
6.2 Additional Checks to Guard Against Systemic Data Loss
6.3 Determining the Calibration Scale and the Verification Points
6.4 Warm-up Time and Stability of Indicated Concentration
6.5 Timing and Frequency of Verifications and Reverifications
6.6 Protecting Transfer Standards Against Damage
6.7 Maintaining Backup Verified Transfer Standards
6.8 Preventive Maintenance
6.9 Transfer Standard Repairs
6.10 Spare Parts Inventory
6.11 Zero-air Supply
6.12 Excess Flow
6.13 Protecting Internal Components of Transfer Standards
6.14 Acceptable Materials
6.15 Line Conditioning and Line Length
6.16 Acceptable Testing Devices
6.17 Data Handling and Documentation
6.18 Use of Standard Forms and Logbooks
6.19 Work Area Cleanliness and Organization
6.20 Laboratory Conditions (Temperature, Relative Humidity)
6.21 Use of Data Loggers or other Automation Software to Run Verifications
6.22 Tools
6.23 Safety
Additional References
22
22
23
23
23
25
27
27
29
31
32
33
33
34
34
34
35
36
36
36
37
37
37
38
38
38
39
39
39
39
41
Appendix A Equations and Example Calculations
1
-------
Equation 1 - Percent difference (% Diff) in measured concentration at the fh concentration point within the ith
cycle 2
Equation 2 - Absolute difference (AbsDiff) in measured concentration at the fh concentration point within the ith
cycle 2
Equation 3 - Linear regression model for the ith cycle (i = 1,2,3) 3
Equation 4 - Least squares estimates for slope and intercept for the fh cycle (i = 1,2,3) 4
Equation 5 - Predicted concentration reported by the candidate transfer standard according to the fitted line
within the ith cycle (i = 1,2,3) 4
Equation 6 - Average slope (m)for a verification test 4
Equation 7 - Average intercept (b)for a verification test 5
Equation 8 - Standard deviation of the three slopes within a verification test (SDm) 5
Equation 9 - Standard Deviation of the three intercepts within a verification test (SDb) 6
Equation 10 - Standard Concentration (ppb) at a given concentration point 7
Appendix B Example Acceptance Testing Data Sheet 1
Appendix C Utilizing NIST 7 Essential Elements of Traceability 1
Appendix D Rationale and Testing Methodology for Os Verification and Reverification Acceptance Criteria
Appendix E Qualification Process
List of Figures
Figure 2-1 EPA Transfer Standard Hierarchy 5
Figure 2-2 O3 Transfer Standard Lifecycle Flowchart 6
Figure 2-3 Standard Reference Photometers 9
Figure 2-4 Verification Frequencies for Bench and Field Transfer Standards 11
Figure 2-5- Traceability Scheme #1: Use of Level 3 Bench Standard and Level 2 Field Standard
13
Figure 2-6 Traceability Scheme #2: Use of Level 3 Field Standard without Level 3 Bench
Standard 15
Figure 2-7 Traceability Scheme #3: Use of Level 2 Bench Standard only 16
Figure 5-1 Verification Procedure Flowchart 24
Figure 5-2 Reverification Procedure Flowchart 26
Figure 6-1 Troubleshooting Flowchart 28
Figure 6-2 Example Verification Points for a Calibration Scale of 180 ppb 32
List of Tables
Table 1-1 List of Major Changes from the Previous Version 3
Table 2-1 List of Standard Reference Photometers in the US 9
Table 4-1 Specifications for Ozone Transfer Standards 19
-------
Table 6-1 List of Common Major and Minor Repairs for O3 Transfer Standards
-------
Preface
Intent of Document
This Technical Assistance Document (TAD) defines, specifies, and formalizes the verification
procedures of NIST-traceable O3 transfer standards used for the calibrations (40 CFR Part 50
Appendix D) and Measurement Quality Checks (40 CFR Part 58 Appendix A) of O3 analyzers.
This document identifies what is necessary to establish and maintain the traceability of O3
measurements within a monitoring network. All O3 transfer standards must meet these
verification requirements. This document updates and replaces the 2013 document entitled
"Transfer Standards for the Calibration of Ambient Air Monitoring Analyzers for Ozone"1 and
the 1979 document entitled "Technical Assistance Document for the Calibration of Ambient
Ozone Monitors"2. This document removes methods no longer in use and updates and
standardizes definitions and procedures. This revision also updates the formatting consistent
with other Quality Assurance documents published by the US Environmental Protection Agency
(EPA) Office of Air Quality Planning and Standards (OAQPS). Additional guidance and
interpretation are available through the contacts listed below or through any EPA office. EPA
strongly encourages those who are implementing O3 traceability programs to request assistance
in planning and implementation.
With this TAD, EPA is detailing its O3 traceability scheme and more specifically defining the
requirements for establishing and maintaining the traceability of O3 measurements within a
monitoring network.
Document Review and Distribution
The information in this document was developed by the members of the Ozone Traceability
Workgroup. The workgroup membership consists of all the Level 1 (Standard Reference
Photometers) O3 standard operators in the United States. Therefore, membership includes EPA
OAQPS, EPA Office of Research and Development (ORD), ten EPA Regional Offices, the
California Air Resources Board (CARB) and the National Institute of Standards and Technology
(NIST). This document has been subjected to a peer review by EPA and the ambient air
monitoring community nationwide; and has been distributed by OAQPS Quality Assurance staff
to promote consistency across EPA in establishing and maintaining O3 traceability within an O3
monitoring network. This TAD may be viewed on the Internet and downloaded from the EPA's
Ambient Monitoring Technology Information Center (AMTIC) Homepage3.
1 Transfer Standards for the Calibration of Ambient Air Monitoring Analyzers for Ozone. EPA-454/B-13-004 U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711, October,
2013. http://www3. epa.gov/ttn/amtic/qapollutant. html
2 Paur, R.J. and F.F. McElroy. Technical Assistance Document for the Calibration of Ambient Ozone Monitors.
EPA-600/4-79-057. U.S. Environmental Protection Agency, Research Triangle Park, NC 27711, September,
1979. http://www.epa.gov/ttn/amtic/cpreldoc.htm.
3 https://www.epa.gov/aintic
1
-------
Recommendations for improvement or revisions are always welcome. Comments should be sent
to the Regional Ozone Traceability Workgroup contact identified in the Acknowledgements
Section. The Ozone Traceability Workgroup will meet periodically to discuss any pertinent
issues and proposed changes.
11
-------
Acknowledgements
This document is the product of the combined efforts of the Ozone Traceability Workgroup.
Workgroup membership for purposes of this revision include individuals implementing the Level
1 Standard Reference Photometer program from the EPA Office of Air Quality Planning and
Standards, EPA Office of Research and Development, EPA Regional Offices, the California Air
Resources Board and NIST. The following individuals are acknowledged for their contributions.
Gregory Noah, EPA OAQPS
Joann Rice, EPA OAQPS
Trisha Curran, EPA OAQPS
* Scott Moore, EPA ORD
Timothy Sharac, EPA OAR CAMD
*Chris St. Germain, EPA Region 1
*Avraham Teitz, EPA Region 2
Mustafa Mustafa, EPA Region 2
Loretta Hyden, EPA Region 3
Kia Long, EPA Region 3
Verena Joerger, EPA Region 3
Mike Crowe, EPA Region 4
*Keith Harris, EPA Region 4
Adam Zachary, EPA Region 4
Richard Guillot, EPA Region 4
Justin Coughlin, EPA Region 5
* Scott Hamilton, EPA Region 5
Michael Compher, EPA Region 5
Carrie Cummings, EPA Region 5
*Clarence Jackson, EPA Region 6
Thien Bui, EPA Region 7
* James Regehr, EPA Region 7
Michael Davis, EPA Region 7
* Joshua Rickard, EPA Region 8
Mathew Plate, EPA Region 9
Roseanne Sakamoto, Region 9
Anna Mebust, EPA Region 9
Bilal Qazzaz, EPA Region 9
^Barbara Bates, EPA Region 9
Chris Hall, EPA Region 10
Ranjit Bhullar, CARB
*Louise Sorensen, CARB
*James Norris, NIST
Bob Lordo, Battelle
Cheryl Triplett, Battelle
Ian MacGregor, Battelle
*A11 Level 1 Standard Reference Photometer contacts are in bold.
in
-------
Glossary of Terms
Acceptance testing - Acceptance testing is testing that ensures and documents that an overall
system is operating properly and as designed
Adjustment - A change made to the measurement device (transfer standard or O3 monitor)
resulting in a new internal calibration factor.
Annual - Means 365 days.
Annual Performance Evaluations (APE) - The measurement quality check that is required in
40 CFR Part 58 Appendix A § 3.1.2.
Application - The designation of a specific transfer standard as either a bench or field standard
at a specific point in time.
As-found - This term is used to describe data recorded prior to any adjustment being made or if
an adjustment has not been made the conditions of a device upon receipt.
As-left - This term is used to describe data recorded after an adjustment has been made or if an
adjustment has not been made the conditions of a device when all services have been completed.
Bench standard - A transfer standard that remains stationary. Stationary means the transfer
standard is placed at one location for the entire verification period (i.e., placed at one monitoring
site and used for remote checks or placed in a laboratory to conduct verifications).
Best practice - A procedure that is accepted as being the most correct.
Calibration - The comparison of a measurement standard, instrument, or item with a standard or
instrument of higher accuracy to detect and quantify inaccuracies and to report or eliminate those
inaccuracies by adjustment. Calibration of an ambient air monitoring analyzer adjusts the
analytical response of the analyzer to more closely agree with a measurement standard of higher
accuracy. A calibration is always followed by a verification.
Calibration scale - The term used to indicate the concentration range that the O3 instrument is
calibrated over.
Candidate transfer standard - A transfer standard that has met all qualification and acceptance
testing and is in a verification process.
Concentration points - Refers to the pollutant concentration run during a verification cycle.
Conditioning - This term is used to describe the process of flowing an ozonated sample through
a component of the transfer standard and ancillary equipment.
Cycle (verification cycle) - This term refers to the collective comparison of standard
concentrations of at least 6 concentration points and a zero.
Field standard - A transfer standard that is transported to field sites for use. Transported means
that a transfer standard is moved from location to location for use (i.e., to conduct one-point
iv
-------
quality control checks or annual performance evaluations). If a transfer standard is not placed at
one location as a bench standard, it is a Field standard.
Indicated concentration - The raw concentration that is reported from the front panel,
calibrated analog voltage output, digital output or other data reporting mechanism from a device.
Internal calibration factors - A term used to describe the number assigned within a transfer
standard referencing a mathematical correction of the output concentration internal to the transfer
standard computer. Internal calibration factors are sometimes termed coefficient, slope,
background or offset. This terminology is vendor specific.
Level 1 - EPA's network of Standard Reference Photometers (SRP).
Level 2 - Any transfer standard that is verified against a Level 1 SRP.
Level 3 - Any transfer standard that is verified against a Level 2 transfer standard.
Major repair - A broad term that describes a repair that is completed to a component of an O3
instrument that has a direct effect on the measurement.
Measurement Quality Check - A one-point quality control check or annual performance
evaluation as specified in 40 CFR Part 58 Appendix A.
Minor repair - A broad term that describes a repair that is completed to a component of a O3
transfer standard that does not have a direct effect on the measurement.
Monitoring organization (MO) - This term will be used to identify any tribal, state or local
organization that is implementing an ambient air monitoring program, especially if they are using
the data for comparison to the National Ambient Air Quality Standards (NAAQS).
One-Point Quality Control Check (QC Check) - The measurement quality check that is
required in 40 CFR Part 58 Appendix A § 3.1.1.
O3 Generator - This term is broadly used as a description of a device that creates an 03 sample.
A typical 03 generator consists of a temperature-controlled chamber where a stable flow of zero
air is pushed across a UV lamp to output ozone concentrations. The amount of ozone generated
can be changed by varying the UV lamp intensity, or zero air flow rate.
QA Handbook - OA Handbook for Air Pollution Measurement Systems, Volume II Ambient Air
Quality Monitoring Program (EPA-454/B-17-001), commonly known as the "QA Handbook".
The most recent version of this document is available on the AMTIC home page. 4
Qualification testing - Initial testing of an overall device design to determine reliability over a
range of variables.
Quality Assurance Project Plan (QAPP) - A tool required by regulation for project managers
and planners to document the type and quality of data needed for environmental decisions and to
describe the methods for collecting and assessing those data. See EPA Requirements for Quality
4 https ://www3. epa. gov/ttn/amtic/qali st. html
v
-------
Assurance Project Plans for Environmental Data Operations (EPA QA/R5) for more
information. 5
Reverification - A verification of a transfer standard that occurs after a verification and when all
acceptance testing passes at a prescribed frequency.
Standard concentration - The concentration used when conducting calibrations, measurement
quality checks or verifications. This concentration is derived from using Equation 10 in
Appendix A and/or using the indicated O3 concentration of the transfer standard. The method
will depend upon the traceability scheme.
Standard Operating Procedure (SOP) - SOPs are written documents that detail the method for
an operation, analysis, or action with thoroughly prescribed techniques and steps, and are
officially approved as the method for performing certain routine or repetitive tasks. See the QA
Handbook and Guidance for the Preparation of Standard Operating Procedures EPA QA/G-6. 6
Transfer Standard - A transportable device or apparatus which, together with associated
operational procedures, is capable of accurately reproducing pollutant concentration standards or
of producing accurate assays of pollutant concentrations which are quantitatively related to an
authoritative standard. In this document this term will be used to describe field and bench O3
transfer standards.
Validation template -The validation template is a table consisting of three criteria: critical,
operational, and systematic criteria, where each criterion has a different degree of implication
about the quality of the data. This table will list the requirements for collecting valid O3 data.
See QA Handbook Appendix D.
Verification - The process used to relate a candidate transfer standard output to a standard of
higher authority.
Verification range - This term is used to describe the extent of concentrations where a
verification is valid. The concentration range is defined by the area between zero and the highest
concentration point in the verification.
Warm-up time - This term is used to describe the amount of time it takes for an instrument to
meet all operational parameters in the instrument manual.
Zero-air - This term is meant to describe the air used to generate a test atmosphere and the air
used to provide a reference measurement for the assay of the sample. 40 CFR Part 50 Appendix
D § 4.4.1 defines Zero-air as follows: "The zero air must be free of contaminants which would
cause a detectable response from the 03 analyzer, and it must be free of NO, C2H4, and other
species which react with 03 " The NIST Specifications for the Standard Reference Instrument
Series 6008, Ozone Standard Reference Photometer defines zero-air as ".. .clean, dry air.
5 https://www.epa.gov/qualitv/epa-qar-5-epa-requirements-qualitv-assurance-proiect-plans
6 https://www.epa.gov/qualitv/guidance-preparing-standard-operating-procedures-epa-qag-6-
march-2001
vi
-------
Preferably, zero-air containing no significant impurities, having less than 1 ppm (parts per
million) total hydrocarbons by volume, and containing 20 - 21% oxygen."
vii
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
1.0 Introduction
1.1 The Importance of O3 Measurement Traceability
Monitor calibrations and measurement quality checks require the generation and assay of known
O3 concentrations from transfer standards that meet the specifications in this TAD. Gaseous O3
standards cannot be stored for any practical length of time due to the reactivity and instability of
the gas. Therefore, traceable O3 concentrations must be generated and "verified" locally.
Devices are commercially available which meet this need. However, these devices must be
calibrated and verified against more authoritative standards prior to being used as a standard
themselves. These standards are called "transfer standards". A transfer standard is defined as a
device or apparatus which, together with associated operational procedures, is capable of
accurately reproducing traceable pollutant concentrations or of producing traceable assays of
pollutant concentrations which are quantitatively related to a higher level and more authoritative
standard. The transfer standard's purpose is to transfer the traceability of a pollutant standard to a
remote point where it is used to verify or calibrate an air monitoring analyzer. The specific
procedures practiced define the resultant measurements' traceability and have a direct effect on
the measured O3 concentrations.
1.2 Code of Federal Regulations (CFR) Requirement7
EPA's Data Quality Objective (DQO) for O3 is stated in 40 CFR Part 58 Appendix A § 2.3.1.2:
Measurement Uncertainty for Automated O3 Methods. The goal for acceptable
measurement uncertainty is defined for precision as an upper 90 percent confidence limit
for the CV of 7 percent and for bias as an upper 95 percent confidence limit for the
absolute bias of 7 percent.
40 CFR Part 58 Appendix A § 2.6.2 provides that "Test concentrations for O3 must be obtained
in accordance with the ultraviolet photometric calibration procedure specified in appendix D to
Part 50 of this chapter and by means of a certified NIST-traceable O3 transfer standard"
(emphasis added). Additionally, 40 CFR Part 50 Appendix D § 4.2 provides that "Transfer
standards must meet the requirements and specifications set forth in Reference 12" (reference 12
is the 2013 version of this document). This document provides updated specifications and best
practices for organizations using O3 transfer standards for NAAQS comparison to demonstrate
NIST traceability.
1.3 Summary of Photometric Technique
The UV photometric technique requires a stable O3 generator, a UV photometer, and a source of
clean, dry, pollutant-free air. A flowing (dynamic) system is set up in which clean air is passed
through the O3 generator at a constant flow rate and discharged into a multiport manifold. The
7 https://www.ecfr.gov/
1
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
O3 concentration in the manifold is assayed by the photometer and is available for calibration of
O3 monitors or verification of transfer standards. The air flow and/or the O3 generator can be
adjusted to provide the approximate O3 concentration desired. The UV photometer is then used
to measure the UV absorption of the generated concentration at a wavelength of 254 nm
according to 40 CFR Part 50, Appendix D § 4.3.1. This transmittance measurement, together
with the well-established absorption coefficient of O3 at that wavelength and various instrument
parameters, is used to calculate the O3 concentration by means of the Beer-Lambert absorption
law. The accuracy of the photometer and adherence to standard operating procedures is critically
important to this technique. Many commercially available instruments have been designated as
Federal Equivalent Methods (FEM)8 and are therefore approved, under the stated conditions, to
be used for the measurement of O3 in ambient air and subsequent comparison to the NAAQS.
The FEM designation only applies to analyzers used for this purpose and does not apply to
transfer standards. The methodology and specifications for O3 transfer standards are in this
TAD.
1.4 O3 Technical Assistance Document History
The original analytical procedure prescribed by the EPA for verifying local O3 concentrations
was a wet chemical technique based on spectrophotometric analysis of iodine generated by O3 in
neutral buffered potassium iodide (NBKI) and referenced to an arsenious oxide primary
standard. EPA amended these regulations by replacing the NBKI technique with more effective
technique based on absorption of ultraviolet (UV) radiation and referenced to the well-
established absorption coefficient of O3 at a wavelength of 254 nm.
On February 8, 1979 (Federal Register, 44:8221-8233), the EPA amended 40 CFR Part 50,
Appendix D to prescribe a calibration procedure for the calibration of reference methods for
measuring O3 in the atmosphere. The procedure is based on the use of UV photometry as the
authoritative standard for O3 and allows the use of transfer standards for the calibration of
ambient O3 monitors, provided such transfer standards are adequately referenced to a UV O3
standard of higher authority (level) and traceability.
To support the implementation of the new UV transfer standard technique, EPA published two
TADs in September 1979. The first TAD was entitled Transfer Standards for Calibration of Air
Monitoring Analyzers for Ozone (EPA-600/4-79-056) with revisions occurring in 2010 (EPA-
454/B-10-001) and 2013 (EPA-454/B-13-004). The second TAD was entitled Technical
Assistance Document for the Calibration of Ambient Ozone Monitors (EPA-600/4-79-057).
While the principles in EPA-600/4-79-056 (and revisions) and EPA-600/4-79-057 remain
8 https://www.epa.gov/amtic/air-monitoring-methods-criteria-pollutants. Federal equivalent method
(FEM) is defined in 40 CFR 58.1 to mean "a method for measuring the concentration of an air pollutant in the
ambient air that has been designated as an equivalent method in accordance with part 53 of this chapter; it does
not include a method for which an equivalent method designation has been canceled in accordance with §
53.11 or § 53.16."
2
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
correct, many of the procedures in those documents are outdated. Therefore, EPA-600/4-79-056
(and revisions) and EPA-600/4-79-057 are being superseded by this document.
1.5 Document Organization
This document focuses on providing practical information and procedures for establishing and
maintaining the traceability of O3 measurements within a monitoring network. This revision also
aims to align the document's formatting with other QA documents published by OAQPS.
The major part of the document is devoted to the specifications and practical procedures
necessary to establish and maintain NIST traceable O3 transfer standards. The beginning
sections provide background and overarching information about traceability and the later
sections provide the technical details needed to successfully implement a traceability program.
Section 4.0 gives the O3 Transfer Standard Requirements and Specifications Table which is
an easy to read table of requirements. This table also provides quick references to sections
within the TAD and outside references. Diagrams and flowcharts have been inserted where
appropriate to illustrate complicated processes. Section 6.0 lists operational considerations as
practical information to be used when working with all O3 transfer standards. Appendix A
provides all equations and example calculations.
1.6 Summary of Major Changes from the Transfer Standards for Calibration of Air
Monitoring Analyzers for Ozone (2013)
This document supersedes the 2013 version of Transfer Standards for Calibration of Air
Monitoring Analyzers for Ozone (EPA-454/B-13-004). The following table is a concise list of
the major changes in this TAD. This table does not encompass all changes.
Table 1-1 List of Major Changes from the Previous Version
2013 Version (EPA-454/B-13-004)
2023 Version (EPA-454/B-22-003)
Transfer standard nomenclature was
ambiguous to application and distance from
Level 1 (i.e., all Level 2 were labelled "bench
standards")
Transfer standard nomenclature is clarified
and based on application (bench standard or
field standard) and distance from Level 1
Reverification frequency was based on
distance from SRP
Reverification frequency is based on
application (bench versus field)
Levels had differing acceptance criteria
Levels 2 and 3 have the same acceptance
criteria
Relative Standard Deviation was used as a
measure of transfer standard verification and
reverification cycles stability
1 Standard Deviation is used for verification
cycles stability. Change from previous
regression slope and intercept are used for
reverification stability.
Six Cycles were required for verifications and
cycles had to be on different days
1 Three stable cycles are required for
verification and can be conducted on same
day
3
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
2013 Version (EPA-454/B-13-004)
2023 Version (EPA-454/B-22-003)
Allowed generator-only devices to be used as
03 transfer standards
Does not allow generator-only devices to be
used as 03 transfer standards
Level 4 transfer standards were allowed
Level 4 transfer standards are strongly
discouraged and are allowed only if additional
requirements are met
Did not include best practices
Includes hands-on operational best practices
for working with O3 transfer standards
Appendix D of this TAD provides descriptions and EPA's rationale for these changes and other verification and
reverification requirements.
This TAD, Transfer Standards for Calibration of Air Monitoring Analyzers for Ozone EPA-
454/B-22-003, clarifies what is needed to establish and maintain the traceability of O3
measurements within a monitoring network and, in some cases, provides flexibilities that were
not available in the previous version. It supersedes EPA-600/4-79-056 (and revisions) and EPA-
600/4-79-057. The Quality Assurance Handbook for Air Pollution Measurement Systems
Volume II9 provides additional information regarding regulatory monitoring for O3. Where
provisions in this technical assistance document have changed (i.e. calling for 3 cycles instead of
the traditional "6X6") or are clarified, implementation will occur over a phase in period of not
more than 2 years from the time this document becomes final.
1.7 Use of the Terms Must and Should
In order to distinguish requirements from provisions that are recommended or a best practice, the
following terms will be used with consistency.
must- when the element is required because it is necessary to establish and maintain the
traceability of O3 measurements
should- when the element is recommended or is a best practice. This term is used when
extensive experience in monitoring provides a recommended procedure that would help establish
or improve the quality of data or a procedure. The process that includes the term is not required
but if not followed, an alternate procedure should be developed that meets the intent of the
procedure. In order to distinguish provisions that are necessary to establish and maintain the
traceability of O3 measurements from those that are recommended or a best practice, the
following terms will be used with consistency.
2.0 EPA's O3 Traceability Scheme
2.1 Description of EPA's O3 Traceability Scheme
NIST traceable O3 transfer standards are required to conduct calibrations and measurement
quality checks of O3 analyzers used to collect data for NAAQS comparison. Gaseous O3
standards cannot be stored for any practical length of time due to the reactivity and instability of
9 https ://www3. epa. gov/ttn/amtic/qali st. html
4
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
the gas. Therefore, O3 concentrations must be generated and verified on site. When the monitor
to be calibrated is located at a remote monitoring site, it is necessary to use a transfer standard
that is traceable to a more authoritative standard. Traceability is the "property of a measurement
result whereby the result can be related to a stated reference through a documented unbroken
chain of calibrations, each contributing to the measurement uncertainty". EPA broadly uses the
NIST "7 Essential Elements of Traceability" as described in Appendix C of this TAD as a guide
to establish and maintain NIST traceability of O3 measurements.
EPA's traceability scheme is carefully designed so that O3 data collected while following the
procedures in this TAD meet the DQO.
Figure 2-1 illustrates EPA's transfer standard hierarchy for O3 measurements. The family of
SRPs are defined as Level 1 transfer standards. Beyond the SRPs, all levels of standards are
numbered (starting with Level 2) based on its "distance in the traceability chain" from a
verification against a Level 1 standard.
For Levels 2 and 3, the transfer standard is first precisely related to a standard one level above by
careful comparison as
described in this TAD.
Traceability is established
by generating known
concentrations of O3 and
concurrently assaying
these with a verified
transfer standard of higher
authority. These data are
used to establish a linear
regression relationship
between the candidate
transfer standard and the
standard one level above
it in the hierarchy.
Provided the acceptance
criteria (listed in Table
4-1 below) are met, the
candidate transfer Figure 2-1 EPA Transfer Standard Hierarchy
standard becomes a
verified transfer standard one level below the level of its predecessor. An unbroken chain of
calibrations is established by either adjusting the internal calibration factors prior to the
verification of the transfer standard being verified, and/or by using Equation 10 of Appendix A
to mathematically calculate the standard concentration from the verified device when
subsequently used. This process appropriately relates the new transfer standard indicated
concentration to its predecessor.
5
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
After a passing verification, the transfer standard may then be used to conduct calibrations or
measurement quality checks. Transfer standards that do not meet the acceptance criteria in Table
4-1 cannot be used until corrective action is taken and a passing verification is completed.
Reverifications of all transfer standards must be conducted at specified intervals. O3 transfer
standards are complex systems consisting of devices that generate and assay O3 concentrations.
Consequently, their verification and use must be in accordance with these prescribed procedures.
This re-comparison serves to ensure that transfer standards are maintaining functionality and
traceability. Table 4-1 also provides a summary of the verification and reverification
frequencies.
2.2 Lifecycle of O3 Transfer Standards
Figure 2-2 O3 Transfer Standard Lifecycle Flowchart
Successful implementation of an O3 traceability program will require significant understanding
of the procedures involved. The following illustration is meant to assist the user with a visual
representation of the general steps involved in maintaining a current verification of an O3 transfer
standard. Figure 2-2 illustrates the typical lifecycle of an O3 transfer standard.
6
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
® - Beginning in the top of Figure 2-2, an O3 transfer standard must undergo qualification
testing either at the vendor or at the monitoring organization (MO). If the vendor has previously
completed suitable qualification testing and can provide that documentation to the user, it is not
necessary for the MO to also conduct the qualification testing since qualification testing is a test
of the design of the instrumentation and configuration. Details of the qualification testing
process can be found in Section 4.2.1 and in Appendix E.
© - The MO should review all specifications and ensure that an adequate transfer standard is
chosen that meets the needs of the monitoring objective.
© - After qualified devices are chosen, the MO must develop a suitable QAPP and SOP to
document and describe how they will implement this TAD.
© - Acceptance testing is the first step in the verification or reverification process. Acceptance
testing ensures the transfer standard is operating as designed. Section 4.2.2 discusses acceptance
testing and Appendix B contains an example worksheet which may be used to document
acceptance testing. If the acceptance testing fails, the user must determine the cause and correct
the problem before attempting the acceptance test again. This may include sending the device
back to the manufacturer for repair. The user must successfully complete acceptance testing
prior to instrument verification or reverification.
© - After successful completion of the acceptance testing, a verification must be conducted
against a standard of higher authority. All transfer standards must be verified and reverified
according to Table 4-1. Section 5.0 gives general steps for the verification.
© - A successfully verified transfer standard may be used within its' verification frequency.
Users are strongly encouraged to conduct other checks to ensure that the transfer standards have
not drifted. Section 6.2 describes several check methods.
® - The transfer standard must undergo acceptance testing and reverification according to the
frequencies in Table 4-1.
® - The transfer standard must be assessed and repaired if any acceptance testing, verification or
reverification fails.
As mentioned above, for an O3 transfer standard to meet EPA's traceability scheme the criteria in
Table 4-1 must be met and documented. If at any time an O3 transfer standard does not meet the
criteria in Table 4-1 or if the verification period has expired, the instrument is no longer a valid
transfer standard and cannot be used for O3 monitoring calibrations or measurement quality
checks. EPA's traceability scheme does allow flexibility for expired transfer standards to
undergo a reverification if all the criteria in Table 4-1 are met (see Section 2.7) but a transfer
standard may not be used if it has exceeded its verification frequency.
2.3 Level 1 Transfer Standard Scheme
All Level 1 transfer standards within EPA's O3 traceability scheme are SRPs and are bound by
the procedures for the EPA's SRP program. Level 1 transfer standards must only be used in a
7
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
laboratory setting with adequate environmental conditions (Section 6.20) and only be used to
verify or reverify lower level transfer standards. The only exception is when a Level 1 SRP is
used to conduct a comparison against an equal Level 1 SRP to provide verification.
All SRPs are designed and built by NIST according to the design specifications in NISTIR 6963
(Paur et al., 2003). Several upgrades have occurred through the years improving the original
design (Norris et al., 2013, 2004).
These publications and NIST's continuous testing of the SRPs as designed concludes that "data
from the entire SRP network have indicated that the SRP is a reliable and repeatable ozone
standard calibration instrument. Intercomparisons conducted annually over more than two
decades have shown the SRPs to be in agreement to better than 0.5% over the concentration
range 100 ppbv to 1000 ppbv and ±1 ppbv over the concentration range of 0 ppbv to 100 ppbv.
This network has met the needs of the user community responsible for ozone calibration by
providing local access to authoritative standards" (Paur et al., 2003).
Figure 2-3 represents the family of Level 1 SRPs. Traceability of NIST's Level 1 SRPs and
EPA's Level 1 SRPs is achieved by demonstrating traceability to the BIPM.QM-K110. The
BIPM.QM-K1 is the International key comparison for O3 reference standards. Participants of the
BIPM.QM-K1 are National Metrology Institutes (NMIs) or Designated Institutes (Dis) of
Member States of the Convention of the Meter. A DI is an Institute within a country that has
been designated by their NMI as the O3 reference laboratory for that country. NIST is the NMI
for the United States of America. An NMI or DI maintains the O3 standard for their Country and
serves as the traceability link for all O3 measurements within that Country. Each NMI or DI
must have an approved quality system and stated Calibration and Measurement Capabilities
(CMC) for O3. Participation in the BIPM.QM-K1 is then required on specific intervals to show
proof of their CMC claims.
Bureau International des Poids et Measures (BIPM)11 and NIST both maintain several Level 1
SRPs. Currently, BIPM SRP #27 serves as the reference standard for the BIPM.QM-K1. BIPM
SRP #28 is always operated concurrently with BIPM SRP #27 as a check of BIPM SRP #27
during all key comparison measurements. These SRP's purpose can change over time if new
SRPs are utilized or other circumstances warrant a change. The results of all participants of the
BIPM.QM-K1 are available on the BIPM website in the Key Comparison Database (KCDB)12.
Successful participation in the BIPM.QM-K1 key comparison is accomplished by demonstrating
agreement to the reference instrument within the stated uncertainties. The acceptance limits of
an EPA Level 1 SRP must be (1.00 ± 0.01 slope and 0.0 ± 1 nmol/mol (ppbv) intercept).
10 https://www.bipm.org/en/bipm/chemistry/gas-metrology/ozone/ozone-comparisons.html
11 https://www.bipm.org/
12 https://www.bipm.org/kcdb/
8
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
EPA ORD LEVEL
BIPM LEVEL 1
NIST LEVEL 1
1 SRP §61
SRP #27
©
SRP #0
©
EPA ORD LEVEL
BIPM LEVEL 1
NIST LEVEL 1
1 SRP #7
SRP §28
SRP #2
EPA ORD LEVEL
:: n • Biannual comparison (Protocol B far the Kef
1 SRP #1
EPA LEVEL 1
SRP #3
EPA LEVEL 1
SRP #4
EPA LEVEL 1
SRP #5
EPA LEVEL 1
SRP #6
EPA LEVEL 1
SRP #8
EPA LEVEL 1
SRP #9
EPA LEVEL 1
SRP #10
EPA LEVEL 1
SRP #13
EPA LEVEL 1
SRP #36
Comparison BIPM.UM-K1, Ozone at ambient level)
Indicates Biannual comparison (Validation of
Standard Reference Photometer, TP 646.0312b)
'i..". ates Annual comparison (SOP for the Verification
and Re-Verification of Eftft's Ozone sup's)
® -NIST participates in the BIPM.QM-K1, ozone at ambient level.
Figure 2-3 Standard Reference Photometers
© - Within EPA, the ORD Metrology laboratory maintains EPA SRP #1, #7 and #61. EPA
ORD sends at least one of these SRPs to NIST for verification biannually. Additional
comparisons are made between SRP #1, SRP #7 and SRP #61 as needed throughout the year.
® - The ORD Level 1 SRP is used to conduct an annual verification against the EPA Regional
SRP's.
In 2020, a total of 64 SRPs exist worldwide. Table 2-1 shows a listing of all SRPs that support
the NAAQS monitoring network within the US.
Table 2-1 List of Standard Reference Photometers in the US
SRP#
Completion Date
Location
Organization
0
August 1985
Gaithersburg, MD
NIST
(backup/traveling)
1
February 1983
RTP, NC
EPA ORD
2
February 1983
Gaithersburg, MD
NIST (National
Standard)
3
August 1983
Edison, NJ
EPA Region 2
4
September 1983
Sacramento, CA
California ARB
5
March 1985
Houston, TX
EPA Region 6
6
March 1985
Chicago, IL
EPA Region 5
7
January 1986
RTP, NC
EPA OAQPS
8
February 1986
Golden, CO
EPA Region 8
9
May 1987
Chelmsford, MA
EPA Region 1
10
November 1987
Athens, GA
EPA Region 4
13
January 1989
Kansas City, KS
EPA Region 7
36
August 2004
San Francisco, CA
EPA Region 9
61
December 2017
RTP, NC
EPA ORD/OAQPS
9
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
2.4 Defining the Terms Level 2 and Level 3 Transfer Standard
The function of Level 2 and Level 3 transfer standards is to duplicate and distribute O3
atmospheres traceable to a Level 1 SRP with a stated uncertainty. Measurement uncertainty
increases with the distance from the first standard in the traceability hierarchy. In traditional
terminology, Level 2 transfer standards were termed "primary standards" and Level 3 transfer
standards were termed "calibrator", "field standard" or "transfer standard" (in either case the
device that is taken to air monitoring sites). The level number (1, 2 or 3) describes the devices
distance from the first standard in the traceability hierarchy whereas the terms "primary
standard" or "calibrator" is ambiguous to that important and meaningful fact. Therefore, EPA is
now exclusively using the terms "Level 1", "Level 2" or "Level 3" in order to give these terms
more meaningful definitions.
2.5 Defining the Terms Bench and Field Transfer Standard
Bench standards are transfer standards that remain stationary. Stationary means the transfer
standard is placed at one location for the entire verification period (i.e., placed at one monitoring
site and used for remote checks or placed in a laboratory to conduct verifications). To ensure
proper functioning of the bench standard, the area where bench standards are operated should be
free of dust, maintain regulated humidity (Section 6.20), have a stable power supply, and suitably
ventilated. Bench standards are only transported when it must be verified or if repairs or
maintenance is necessary.
Field standards are transfer standards that are transported to field sites for use. If a transfer
standard is not placed at one location as a bench standard, it is a field standard. For example,
calibrators designated to perform annual performance evaluations (APE) are field standards
because they must travel from site to site to conduct audits. Field standards should always be
handled with care to prevent damage.
A Level 2 or 3 transfer standard's application may be either as a bench or field standard. All
Level 1 SRPs must function as bench standards. A bench standard may become a field standard
at any time; however, the standard would require more frequent reverifications as a result. A
field standard cannot become a bench standard until it is appropriately verified. For example,
consider a standard which is verified against a Level 1 SRP and then designated as a Level 2
bench standard used to conduct Level 3 verifications or Level 2 reverifications. Anytime during
its verification period, this Level 2 bench standard may be re-designated as a Level 2 field
standard and transported to monitoring sites to conduct work (i.e., calibrations or measurement
quality checks). However, if this happens, the standard cannot be re-designated back to a Level
2 bench standard until it is verified against a Level 1 SRP.
2.6 Bench versus Field Transfer Standard Verification Frequency
In previous versions of this TAD, EPA stated verification frequencies that were based on the
distance the device was from the Level 1. Generally, the further away from the Level 1 the more
verifications were required to maintain traceability. Level 2 verifications were required
10
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
annually, and Level 3 verifications were required every 6 months. In this revision, EPA uses the
application (bench or field standard) of the transfer standard as a means of determining
verification frequencies. Field standards are more likely to experience drift due to vibrations and
other impacts from repeated transport; therefore, field standards should be verified and checked
more frequently. Bench standards will experience less of the impacts related to transport and will
be much less likely to experience performance shifts; therefore, bench standards require less
frequent checks.
Figure 2-4 shows the verification frequencies and traceability hierarchy EPA has defined for the
bench and field standards. Table 4-1 provides a complete listing of O3 transfer standard
verification frequencies and requirements. It should be noted that this diagram shows the
verification and reverification frequencies for all possible combinations of bench and field
transfer standards. Two Level 2 and Level 3 transfer standards (one bench and one field) are not
required. See Section 2.8 for details on traceability schemes. Bench standards require an annual
verification frequency and field standards require a 6-month verification frequency as described
below.
LEVEL 1 SRP
All Ltvtf 2 SfaRtiafds. i» vMffifid to in SRP aMUttliy
*L2 BENCH
Stationary
**L2 FIELD
Transportable
(?)
**L3 FIELD
Transportable
*L3 BENCH
Stationary
•All Bench standards must be verified annually.
**Ali Field standards must be verified every 6 months.
indicates Verification
i ndicates Reverification
Figure 2-4 Verification Frequencies for Bench and Field Transfer Standards
® - All Level 2 transfer standards must be initially verified against a Level 1 SRP, then verified
annually or upon failure of a reverification, each time against a Level 1 SRP.
© - Level 2 field standards must be reverified against a standard of equal or higher authority
(Level 2 bench standard or Level 1 SRP) every 6 months.
11
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
® - All Level 3 transfer standards must be initially verified against a Level 2 bench standard.
© - Level 3 field standards must be reverified against a Level 2 bench standard every 6 months.
© - Level 3 bench standards must be reverified against a standard of equal or higher authority
(Level 2 bench standard or Level 2 or 3 field standard) annually.
Users are cautioned that drift can occur in O3 transfer standards. Calibrating an O3 monitor with
a transfer standard that has unknowingly drifted can cause systemic data loss in a monitoring
network. Therefore, EPA strongly recommends that additional checks be implemented to
determine whether drift has occurred in the transfer standard. Section 6.2 provides additional
checks recommended by EPA to guard against data loss.
A verification timeframe begins on the day the verification is completed and not the day the
transfer standard is put into service. This includes both newly verified transfer standards as
well as reverifications. For example, if a transfer standard is verified on January 1, is placed on
the equipment shelf and not used until May 1 then the reverification timeframe begins on
January 1 and not May 1. This also applies to all field standards. If a bench standard is re-
designated as a field standard, then a 6-month reverification is required (i.e., 6 months from the
time the bench standard was verified).
2.7 Lapses in Verification Frequency
A transfer standard may not be used after the reverification frequency time period has been
exceeded. For example, if a Level 3 field transfer standard was verified on January 1 of a given
year, it may not be used after July 1 of that same year. The intent of defining frequencies for
reverifications is to ensure that drift has not occurred in the measurement system and therefore
ensure that drift is not passed down through the traceability hierarchy to the monitor reporting
the ambient O3 concentration. Frequent checks provide assurances that the system is operating
as designed and that the stated output is correct. It is important to consistently conduct checks at
these defined frequencies. However, EPA recognizes that solely because a transfer standard has
exceeded its verification frequency this in and of itself does not warrant a new verification (all
three cycles). A transfer standard that has narrowly exceeded its verification time period (i.e., by
30 days) may be reverified if all acceptance testing criteria are met and the device is operating
normally. The reverification data must be reviewed against the acceptance criteria and especially
against the previous verification data to be certain only acceptable drift has occurred in the
measurement system. This process must be fully documented in the QAPP and SOP.
2.8 Level 2 and 3 Transfer Standard Scheme
The primary function of Levels 2 and 3 transfer standards is to duplicate and distribute traceable
concentration standards to O3 monitoring sites where traceability to a Level 1 transfer standard is
required. Traceability is required for all O3 monitors that compare their data to the NAAQS.
Depending on the size and complexity of the MO, various levels and numbers of transfer
standards are needed; and many different scenarios will be acceptable. A small MO with only
12
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
one or two sites may be able to position a Level 2 transfer standard at each site. This
arrangement will not be practical for a large monitoring network since each Level 2 transfer
standard requires an annual verification against a Level 1 SRP. Larger networks might benefit
from bringing two or more Level 2 transfer standards for an annual verification against an SRP.
Three of the most common schemes are described below. The quantity and type of transfer
standards needed at Levels 2 and 3 will depend on many factors. Many variations will be
acceptable. Each MO must assess their situation and determine an appropriate
traceability scheme. EPA strongly encourages those who are implementing O3 traceability
programs to request assistance in planning and implementation.13
The circled numbers in the figures correspond to the circled numbers in the descriptions
for each step in the traceability scheme.
2.8.1 Scheme #1 - Use of Level 3 Bench Standard and Level 2 Field Standard
Figure 2-5 shows a simplified roadmap to illustrate the verification and reverification pathways
for Levels 2 and 3 transfer standards in this scheme. This figure shows an example with four
LEVEL 1 SRP
<6>-*
/2f
M* y/
L3 FIELD
#1
L3 FIELD
#2
L3 FIELD
#3
L3 FIELD
#4
L2 NPAP
¦<5>
L3 BENCH
#1
L3 BENCH
#2
L3 BENCH
#3
L3 BENCH
#4
/
O3
MONITOR
» Indicates Verification
*. Indicates Reverification
_ _ ». Indicates Calibration or Quality Control Check
Indicates Independent Audit (Annual Performance
Evaluation orThru-the-Probe)
Figure 2-5- Traceability Scheme #1: Use of Level 3 Bench Standard and
Level 2 Field Standard
Level 3 bench standards and four Level 3 field standards, where the Level 3 bench standards are
positioned at specific monitoring sites and used primarily to perform O3 monitor calibrations
13 For further assistance in planning your transfer standard scheme, contact your Regional EPA office.
13
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
and/or quality control (QC) checks (40 CFR Part 58 Appendix A § 3.1.1), and the Level 3 field
standards are used to perform APEs of O3 monitors according to 40 CFR Part 58 Appendix A §
3.1.2. One advantage of this scheme is that the Level 3 bench standard can remain at a
monitoring site and is not required to be "swapped out" with a newly reverified Level 3 transfer
standard every 6 months. The Level 3 bench standard is reverified at the monitoring site until it
fails to meet the acceptance criteria. One disadvantage of this scheme is that, due to the
infrequent checks against the Level 2 bench standard, vigilance should be used to ensure drift has
not occurred in the transfer standard chain. EPA strongly recommends that additional checks be
implemented to determine if drift has occurred. Section 6.2 provides additional checks
recommended by EPA to guard against data loss.
® - All Level 2 transfer standards are verified against a Level 1 SRP annually (typically 2-3
Level 2 transfer standards are verified). One of the Level 2 transfer standards is designated as a
Level 2 field standard and the remaining Level 2 standard(s) is designated strictly as a Level 2
bench standard and not moved from the laboratory.
© - A Level 2 bench standard(s) is used to verify numerous Level 3 transfer standards. Each
verified Level 3 transfer standard is designated as either a bench or field standard. Frequent
intercomparison between the Level 2 standards is strongly recommended.
© - Level 3 bench standards are deployed to specific monitoring sites and used in a stationary
application to conduct O3 monitor calibrations and/or quality control (QC) checks on O3
monitors. It is also appropriate to conduct O3 monitor calibrations with a Level 2 field standard
at the monitoring site.
© - Level 3 field standards are used to conduct APEs of O3 monitors. The Level 3 transfer
standard used to conduct calibrations and QC checks must be different than the transfer standard
used to conduct the APE.
© - Level 3 bench standards are reverified against Level 2 bench, Level 2 field or Level 3 field
standard annually (at the monitoring site or at the Level 2 bench standard location). EPA
strongly recommends that additional checks be implemented to ensure that drift has not occurred
in the transfer standard.
© - Level 2 field standards and Level 3 field standards are reverified against Level 2 bench
standards every 6 months. EPA strongly recommends that additional checks be implemented to
determine if drift has occurred in the transfer standard.
® - All Level 2 transfer standards are reverified against a Level 1 SRP annually (typically 2-3
Level 2 transfer standards are verified).
® - National Performance Audit Program (NPAP) audits are conducted according to 40 CFR
Part 58 Appendix A § 3.1.3.
14
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
2.8.2 Scheme #2 - Use of Level 3 Field Standard without Level 3 Bench Standard
Figure 2-6 shows a simplified roadmap to illustrate the verification and reverification pathway
for Levels 2 and 3 transfer standards in this scheme. Note that all Level 2 standards are
designated as bench standards. One advantage of this scheme is that the Level 3 field standards
must be brought back to the laboratory hosting the Level 2 bench standard for a direct
comparison, thereby adding additional assurance that drift has not occurred in the transfer
standard. One disadvantage of this scheme is that Level 3 field standards are transported from
site to site which subjects transfer standards to shock, vibration and temperature extremes which
can cause measurement drift. EPA strongly recommends that additional checks be implemented
to ensure that drift has not occurred. Section 6.2 identifies additional checks recommended by
EPA to guard against data loss.
©
D
Indicates Verification
Indicates Reverification
». Indicates Calibration or Quality Control Check
Indicates Independent Audit (Annual Performance
Evaluation or Thru-the-Probe)
Figure 2-6 Traceability Scheme #2: Use of Level 3 Field
Standard without Level 3 Bench Standard
® - All Level 2 transfer standards are verified against a Level 1 SRP annually (typically 2-3
Level 2 transfer standards are verified).
© - The Level 2 bench standard(s) is used to conduct verifications and reverifications of Level 3
field standards.
15
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
® - Level 3 field standards are used to conduct O3 monitor calibrations and/or QC checks (40
CFRPart 58 Appendix A § 3.1.1) on O3 monitors.
© - Level 3 field standards are used to conduct APEs of O3 monitors according to 40 CFR Part
58 Appendix A § 3.1.2. The Level 3 transfer standard used to conduct calibrations and QC
checks must be different than the transfer standard used to conduct the APE.
© - All Level 2 transfer standards are reverified against a Level 1 SRP annually (typically 2-3
Level 2 transfer standards are verified).
© - National Performance Audit Program (NPAP) audits are conducted according to 40 CFR
Part 58 Appendix A § 3.1.3.
2.8.3 Scheme #3 - Use of Level 2 Bench Standard only
Figure 2-7 shows a simplified roadmap to illustrate the verification and reverification pathway
for Level 2 transfer standards in this scheme. Here, the single Level 2 bench standard is located
at a monitoring site. One advantage of this scheme is that the Level 2 transfer standard is used to
directly calibrate the O3 monitor at the monitoring site, which means the reported O3
concentration is one level closer to the Level 1 SRP than in Schemes #1 and #2. One major
disadvantage of this scheme is that, due to the infrequent checks of the Level 2 bench standard,
and the isolation of that standard at the monitoring site (no additional checks from independent
transfer standards) vigilance must be used to ensure drift has not occurred in the transfer standard
chain. Another disadvantage is that access to a Level 1 SRP is limited and it might not be
practical to provide Level 2 field standards at each monitoring site. EPA strongly recommends
©
Indicates Verification
Indicates Bewrificition
»> Indicates Calibration or Quality Control Check
Indicates independent Audit (Annual Performance
Evaluation or Tfiru-the-Probe)
Figure 2-7 Traceability Scheme #3: Use of Level 2 Bench Standard
only
INDEPENDENT
STANDARD
16
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
that additional checks be implemented to ensure that drift has not occurred. Section 6.2 provides
additional checks recommended by EPA to guard against data loss.
© - All Level 2 transfer standards are verified against a Level 1 SRP annually. The number of
Level 2 transfer standards depends on the number of monitoring sites.
© - The Level 2 bench standard(s) is placed at the monitoring site and used to conduct
calibrations or QC checks (40 CFR Part 58 Appendix A § 3.1.1) on O3 monitors.
© - An independent transfer standard is used to conduct APEs according to 40 CFR Part 58
Appendix A § 3.1.2.
© - All Level 2 transfer standards are reverified against a Level 1 SRP annually (typically 2-3
Level 2 transfer standards are verified).
© - National Performance Audit Program (NPAP) audits are conducted according to 40 CFR
Part 58 Appendix A § 3.1.3.
2.9 Quality Documentation
Quality system requirements are discussed in EPA documents CIO 2105.0 (formerly EPA Order
5360.1 A2)14, EPA QA/R-2, EPA Requirements for Quality Management Plans15, and EI ¦'A
QA/R-5, EPA Requirements for Quality Assurance Project Plans16. Traceability and transfer
standards must be supported by trained operators and documented procedures. For a claim of
traceability to be made, all users of all Levels of O3 transfer standards must incorporate all
relevant information into their Quality Assurance Project Plan (QAPP) and Standard Operating
Procedures (SOP). QAPPs and SOPs must provide enough detail to describe their specific
traceability scheme with adequate detail and should follow the guidelines in the QA Handbook
for Air Pollution Measurement Systems, Volume II Ambient Air Quality Monitoring Program
(EPA-454/B-17-001), commonly known as the "QA Handbook".17
This TAD provides best practice procedures while also allowing flexibility for Mos to implement
traceability of O3 measurements according to their specific situation. In any case, it is the MO's
responsibility to ensure that their quality system is documented, accurate, and implemented.
EPA Regions and Mos using O3 transfer standards should incorporate procedures consistent with
those outlined in this document into their quality system documentation.
2.10 Training
Training is a required element of an air monitoring QAPP, which includes training plans for
technical staff. Minimum training requirements should be developed, completed and
documented so that users understand the equipment and procedures. Training should include
14 htjps: //www,epa myjs j Jeg/Eroducji onMes/20j^^
15 https://www.epa.gov/qnalitv/epa-qar-2-epa-reqnirements-qnalitv-management-plans
16 life:/fegw
17 https://www3.epa.gov/ttii/aintlc/qalist.htinl
17
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
reviewing the QAPP, SOP, relevant instrument manual(s) and hands-on training or shadowing
with an experienced technician. Mos should incorporate this training in a comprehensive plan
for all personnel involved in O3 monitoring work. Training may also be available from vendors
and can be negotiated when making purchases of commercially available equipment.
3.0 Types of O3 Devices
3.1 Devices with Photometers
All O3 transfer standards must meet the specifications in 40 CFR Part 50 Appendix D § 4.3 and
Table 4-1 of this TAD. Users should contact their Regional EPA office prior to putting into
service any device that is not specifically designed by the manufacturer as an O3 transfer
standard.
3.2 Generator-Only devices
Many commercially available O3 analyzers can be purchased with an internal zero-air scrubber,
O3 generator and a valve that can be used to introduce a zero and O3 test atmosphere to the
photometer at a preprogrammed time. This is commonly referred to as an internal O3 generator,
internal ozonator or internal zero/span option in an O3 monitor. These devices do not contain an
independently verified photometer and cannot be used to conduct measurement quality checks
(required by 40 CFR Part 58 Appendix A) or calibrations. The appropriate use for a generator-
only device is to conduct automated zero/span checks (usually at midnight each night).
Information from these devices should be charted and used to assess the relative stability of the
photometer and the general trend of the responses. This data should be used as a diagnostic
check to inform the user whether an action limit is breached thereby triggering an evaluation and
possible recalibration of the monitor prior to the point of data loss.
The user should set up the nightly check to run automatically and graph the results in a control
chart. The user should then frequently review the general trend of the percent differences. Once
an action limit (a typical action limit is 50% of the acceptance criteria) is reached, the user
should take corrective action. It is also possible to use a data logger or other system to
automatically make this determination and send warnings to alert the MO of a breach in action
limits.
4.0 Verification and Reverification Requirements for O3 Transfer Standards
4.1 O3 Transfer Standard Requirements and Specifications Table
The following table provides a summary of the requirements for O3 transfer standards. The last
column also identifies the section or appendix of this document where specific requirements can
be found and provides citations to additional documents that provide additional relevant
information. A candidate transfer standard must pass all the requirements below before being
used for O3 measurement activities.
18
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Requirement
Frequency
Acceptance Criteria
Reference/Additional
Information
1 .evel
(SUP lo SUP Comparison)1
Qualification
Not Applicable
Not Applicable
(Norris et al., 2013,2004; Paur et al., 2003;
Viallon et al., 2006)
Acceptance Testing
(Pre-Verification SRP
QC checks for
temperature and
pressure circuits)
Prior to Level 1
reverification
Prior to use for
Level 2
verification
All acceptance criteria are found in
SRP/SOP-1
SRP/SOP-1 Section 9.5.2, 9/18/2015
Stability Monitor
Prior to each use
All acceptance criteria are found in
SRP/SOP-1
SRP/SOP-1 Section 9.5.2.10, 9/18/2015
Verification
(involves 6 cycles)
Upon receipt and
upon major
repair
Regression slope = 1.00 ± 0.01
Regressions intercept < ±1 ppb
SRP/SOP-1 (9/18/2015), Section 1.3
SRP/SOP-1 (9/18/2015), Section 9.7.20
All Lcvc
2 and 3 Transfer Standards2-'
Qualification
Commercial
devices -
conducted at
vendor
Non-commercial
devices - 1 time
prior to
acceptance
testing
Repeatability within ±4% or 4 ppb
from its indicated value
(whichever is greater)
Section 4.2.1
Appendix E Qualification testing conducted
by the manufacturer if commercial device.
Acceptance Testing
Upon receipt
(new), following
repair, or prior to
conducting
Verifications or
Reverifications
Per manufacturer specifications
Section 4.2.2
Appendix B provides a generic Acceptance
Testing worksheet
Users should also consult the Manufacturer
Manual
Verification
(involves 3 cycles4)
Upon receipt
(new),
adjustment, or
major repair.
All Level 2
transfer
standards must
be verified
annually against
a Level 1 SRP
Each point difference < ± 3.1% (or
±1.5 ppb for concentration points
below 50 ppb)
All Regression slopes = 1.00 ± 0.03
All Regression intercepts = 0 ± 3
ppb
Standard Deviation of the 3 Slopes <
± 0.0075
Standard Deviation of the 3
intercepts < ± 1.00 ppb
Section 4.4.1
Appendix A provides example calculations
and equations.
Level 3 Bench Transfer Standard^
Reverification
(involves 1 cycle4)
Annually
Each point difference <±3.1% (or
±1.5 ppb for concentration points
below 50 ppb)
Regression slope of reverification
cycle must be within ±0.015 of
most recent verification slope
Regression intercept of
reverification cycle must be within
±1.5 ppb of most recent
verification intercept
Section 4.4.2
Appendix A provides example calculations
and equations.
Levels 2 and 3 licit! Trims for Standards0
Reverification
(involves 1 cycle4)
Every 6 months
Same as Level 3 Bench Transfer
Standard Reverification (above)
Same as Level 3 Bench Transfer Standard
Reverification (above)
Table 4-1 Specifications for O3 Transfer Standards
19
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
'Level 1 SRP requirements are outside the scope of this document. Please see the Standard Operating Procedure for the Verification and Re-
Verification of EPA's O3 Standard Reference Photometers for more details.
2Level 2 Transfer standard verification services are only available through the Standard Reference Photometer Program. Contact the Regional
EPA office for scheduling. Small agencies may be able to avoid the cost of Level 2 standard equipment by obtaining access to Level 2 standards
through EPA Regional Offices, State agencies, or other cooperating agencies. Some commercial laboratories may also provide 03 transfer
standard verification services.
Acceptance criteria and verification frequencies listed are minimum requirements. Organizations should review their procedures and assess
their technical expertise with O3 transfer standards and determine if more stringent criteria should be used (i.e., more frequent verifications) to
meet EPA's DQO for 03. Contact the EPA Regional Office for assistance in designing and implementing a 03 traceability scheme.
4 A cycle consists of at least 6 concentrations and zero. Verification points must be chosen according to Section 6.3.
5A Level 3 bench reverification may be conducted against any equal or higher level transfer standard. Additional checks and data reviews are
strongly recommended (See Section 6.2).
6Level 2 and Level 3 field standards must be reverified against a Level 2 bench standard at least every 6 months. All Level 2 transfer standards
must be verified annually against a Level 1 SRP.
4.2 Requirements for all Levels
As described in Section 2.2, the general process for maintaining a verified transfer standard is
qualification, acceptance testing, verification and reverification. Qualification and acceptance
testing ensure the device is properly designed and that it is operating as designed. Verification
and reverification testing are used to relate the candidate standard output to another verified
transfer standard in the hierarchy by concurrently generating and assaying various O3
concentrations.
4.2.1 Qualification Testing Requirements
Qualification is initial testing of an overall device design to determine reliability over a range of
variables. Before a device design can be used as a transfer standard, it must be tested and shown
to have adequate performance and reliability. The design must be repeatable over reasonable
periods of time and over the range of conditions encountered during lab and/or field use and
during transport. Qualification of a transfer standard design may require a series of initial tests
to determine reliability.
All O3 transfer standards must meet the general requirements for qualification. The transfer
standard output should not vary by more than ± 4% or ± 4 ppb (whichever is greater) from its
indicated value over a stated range of any of the conditions to which it might be sensitive. All
devices used as a transfer standard must complete adequate qualification testing according to
Appendix E and the results must be documented.
For most commercially available devices, qualification testing is completed at the vendor during
the design phase of manufacturing and with a representative sample of devices. It is important
that as Mos plan purchases of new generation transfer standards, they consult with the
manufacturers on the processes used to qualify the instruments. It is the user's responsibility to
ensure that qualification testing has been completed for each commercially available device
purchased. Even if design qualifications have been completed at the vendor, there can be value
in conducting qualification testing for newly acquired instruments. This can help ensure the
20
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
instalments will perform under various field conditions that might not be observed during a
traditional verification process.
4.2.2 Acceptance Testing Requirements
Acceptance testing is testing that ensures and documents that an overall system is operating
properly and as designed. Acceptance testing must be completed after a new transfer standard is
received from the manufacturer, prior to verification or reverification of a transfer standard,
when a transfer standard is shipped, or when a device requires repair. For new instruments or
instruments undergoing repair, EPA recommends that the instruments be operated over several
days to ensure that the device is field-ready following any such repair.
Acceptance testing should not be confused with qualification testing (Section 4.2.1) in that it is
primarily an examination of the components in the device. Diagnostics data are readily available
from the device's front panel and should be reviewed and documented frequently. Acceptance
testing should be completed using vigilance but is not meant to be a time-consuming part of the
verification process. However, if any component in the acceptance testing is outside the
manufacturers recommendation then those components must be corrected prior to verification.
An example acceptance test data worksheet is provided in Appendix B.
It is the transfer standard owners responsibility to ensure all acceptance testing and preventive
maintenance is completed and to provide that documentation as part of the verification process.
Elements of a proper acceptance test should include:
1. Documentation of testing date, time, operator, instrument make/model/serial
number (SN);
2. Documentation ensuring routine maintenance required by the instrument manual
has been performed;
3. Direct comparison of sensors impacting the measurement (i.e., sample pressure,
sample temperature, analog outputs) if recommended by the instrument manual or
if sensor adjustment is required;
4. Documentation of diagnostic parameters in the instruments' menu system and
comparison to the manufacturer's specification (i.e., sample pressure, sample
temperature, flow);
5. Review and verify that all acceptance test data are within acceptable limits.
4.3 Level 1 Requirements
Level 1 requirements are not addressed in this TAD. Level 1 transfer standards are addressed in
the most recent version of EPA's SRP/SOP18 and NIST's Gas Sensing Metrology Group Quality
Manual (QM-III-646.03), following TP 646.0312b (Validation of Standard Reference
Photometer).
18 https://www3.epa.gov/ttnamtil/files/ambient/aaac/SRP SQP%20R1%209 .1.8 15.pdf
21
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
4.4 Level 2 and 3 Verification and Reverification Requirements
Verification and reverification procedures must be developed consistent with the specifications
outlined in this TAD. All Level 2 and Level 3 transfer standards require a verification against a
Level 1 SRP or a Level 2 bench standard, respectively. The transfer standard of higher authority
must have a current verification and all verifications and reverifications must be conducted by a
qualified technician. Adjustments to the candidate transfer standards' internal calibration factors
must be made prior to verification. Internal calibration factors must be tracked. No adjustments
are allowed prior to a reverification.
Each testing cycle which consititutes a verification or reverification of a Level 2 or Level 3
transfer standard will be accomplished by generating and simultaneously assaying varying
concentrations of O3 in the range that the device will be used. A linear regression analysis is
conducted using the indicated concentrations of the candidate transfer standard and the standard
concentrations of the transfer standard of higher authority. Care must be taken when reviewing
candidate transfer standard verification and reverification data to ensure that active and excessive
drift is not occurring.
An example verification dataset and calculations can be found in Appendix A.
4.4.1 Verification Requirements
Verifications must be conducted at the frequency stated in Table 4-1. The following is a list
of the verification requirements for Levels 2 and 3 transfer standards.
1. A verification consists of a minimum of three (3) stable testing cycles, each cycle
consisting of a zero and at least six (6) upscale concentration points. Concentration
points must be chosen according to Section 6.3. The following must hold within and
across these testing cycles (equation numbers refer to Appendix A):
a. Each point difference must be < ±3.1% (Equation 1) or ±1.5 ppb for concentration
points below 50 ppb (Equation 2).
b. All Regression Slopes must be 1.00 ± 0.03 (Equation 4).
c. All Regression Intercepts must be 0 ± 3 ppb (Equation 4).
d. Standard Deviation of the 3 Slopes must be < ± 0.0075 (Equation 8).
e. Standard Deviation of the 3 Intercepts must be < ± 1.00 ppb (Equation 9).
4.4.2 Reverification Requirements
Reverifications must be conducted at the frequency stated inTable 4-1. The following is a
list of the reverification requirements for Levels 2 and 3 transfer standards.
1. A reverification consists of a minimum of one (1) stable testing cycle consisting of a
zero and at least six (6) upscale concentration points. Concentration points must be
chosen according to Section 6.3. The following must hold for this testing cycle (equation
numbers refer to Appendix A):
a. Each point difference must be < ±3.1% (Equation 1) or ±1.5 ppb for concentration
points below 50 ppb (Equation 2).
22
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
b. All Regression Slopes must be 1.00 ± 0.03 (Equation 4).
c. All Regression Intercepts must be 0 ± 3 ppb (Equation 4).
d. Regression Slope of reverification cycle must be within ±0.015 of most recent
verification slope
e. Regression Intercept of reverification cycle must be within ±1.5 ppb of most
recent verification intercept
4.5 Level 4 Verification and Reverification Requirements
Level 4 transfer standards (for which a verified Level 3 standard would serve as the transfer
standard of higher authority) are strongly discouraged. However, in the rare instance when a
Level 4 is the only option for establishing traceability, the MO should work with the EPA
Regional office for assistance in developing acceptance criteria and procedures that are adequate
to meet the EPA's DQO for O3 and overall minimize uncertainty. At a minimum, Level 4
transfer standards must be reverified at least quarterly. All other minimum requirements for
Level 4 transfer standards must be at least as stringent as the Level 3 transfer standard criteria
listed inTable 4-1. To guard against systemic data loss, more frequent reverifications or other
cross checks must be completed when using Level 4 transfer standards. Section 6.2 provides
additional checks recommended by EPA to guard against systemic data loss.
5.0 Verification and Reverification Procedure for O3 Transfer Standards
5.1 Verification Procedure
Figure 5-1 gives a generalized flow chart and step-by-step procedure meant to assist the user in
acquiring an overall understanding of the verification process for development of their specific
SOPs. Users must develop detailed SOPs that are inclusive of all steps and describes their
traceability scheme consistent with this TAD.
The circled numbers in the flowchart correspond to the circled numbers in the verification
steps.
23
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Verification
Procedure
Warrr up and
conditio-* csid'date
transfe- standard
Complete
preventive
maintenance
Preventive
maintenance
completed?
Calibration
factors match
previous
verification?
HO
ASSESS
Run
preliminary
zero and high
point
PASS
Run 3 complete
verification
cycles
FAIL
Review ail data
and assess
results
PASS
Save report, affix
summary report to
verified transfer
standard
Figure 5-1 Verification Procedure Flowchart
® - Place the candidate transfer standard as near as possible to the standard of higher authority
to minimize the length of line used - preferably within one (1) meter. Lines must be properly
conditioned according to Section 6.15. Warm-up and condition the candidate transfer standard.
Ensure the transfer standards are not sampling room air.
Steps 2 and 3 can be conducted while the device is warming up.
© -Review all preventive maintenance documentation from the candidate transfer standard.
© - Review the previous verification data, verification history, as-left internal calibration factors
from the previous verification, and the current internal calibration factors. If the internal
calibration factors have changed, determine why they have changed and if the transfer standard
has been used subsequently. Changing the internal calibration factors in a transfer standard
voids the previous verification.
24
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
© - After adequate warm-up, complete all acceptance testing. Document and review all
acceptance testing data to ensure that the candidate transfer standard is operating within the
manufacturer's specifications.
© - Run a preliminary zero and a preliminary high point. Any stable source of ozone may be
used to generate the test concentrations. The high point should be the highest point in the
verification (See Section 6.3). Compare the results against established action limits. If an action
limit is exceeded, then assess and correct the problem. This could include repair, additional
maintenance or adjusting the internal calibration factors. Ensure all parameters are within the
manufacturer's recommendations. If repair or adjustments are made, conduct acceptance testing
(Section 4.2.2).
© - Run at least three (3) complete cycles consisting of six (6) concentration points (additional
concentration points are preferred) and a zero. See Section 6.3 for information on how to chose
verification points.
® - After the appropriate number of cycles have been completed, review the data and verify
they meet all requirements in Table 4-1. All equations, an example verification dataset, and
calculations can be found in Appendix A. If any of the requirements are not met, assess the
problem and begin the verification process from the beginning.
® - Ensure all documentation is complete and all records are saved to the appropriate data
storage system. The MO may place a summary report on the verified transfer standard which
states at a minimum: the date of the verification, the date of verification expiration, name of
person conducting the verification, make/model/SN, current internal calibration factors,
dates/slopes/intercepts of the original verification cycles, the average slope and the average
intercept. If Equation 10 in Appendix A is used in the traceability scheme, then the user should
also state the current equation to be used for calculating the standard concentration when
subsequently used. A summary report should be placed on the top of the transfer standard.
5.2 Reverification Procedure
Figure 5-2 gives a generalized flow chart and step-by-step procedure meant to assist the user in
acquiring an overall understanding of the reverification process for development of their specific
SOPs. The reverification procedure differs from the verification procedure in that only one (1)
cycle is required if all acceptance criteria are met. If at any time during the reverification
process the criteria in Table 4-1 are not met, the candidate transfer standard must be
assessed and if warranted undergo a new verification according to Section 5.1.
25
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Reverification
Procedure
Complete
preventive
maintenance
NO
£
Calibration
factors match
previous
verification?
NO
Assess arid goto
verification
procedure
Section 6,1
ir major repair,
and/er calibration
factors require
adjustment
preliminary
zero and high
point
PASS
Complete 1
verification
cycle
procedure
Section 5,1
t I
FAIL
Review all data
and assess
results
PASS
Save report, affix
summary report to
verified transfer
standard
Figure 5-2 Reverification Procedure Flowchart
® - Place the candidate transfer standard as near as possible to minimize the length of line used -
- preferably within one (1) meter. Lines must be properly conditioned according to Section 6.15.
Warm-up and condition the candidate transfer standard. Ensure the transfer standards are not
sampling room air.
© - Review all preventive maintenance documentation from the candidate transfer standard.
® - Review the verification data, verification history, as-left internal calibration factors from the
verification, and the current internal calibration factors. If the internal calibration factors have
changed, determine why they have changed and if the transfer standard has been used
subsequently. Changing the internal calibration factors in a transfer standard voids the previous
verification and the candidate transfer standard must undergo a new verification according to
Section 5.1.
26
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
© - After adequate warm-up, complete all acceptance testing. Document and review all
acceptance testing data to ensure that the candidate transfer standard is operating within the
manufacturer's specifications.
© - Run a preliminary zero and a preliminary high point. Any stable source of ozone may be
used to generate the test concentrations. The high point should be the highest point in the
verification. Compare the results against established action limits. If an action limit is exceeded,
then assess and correct the problem. This could include a major or minor repair, additional
maintenance and/or adjusting the internal calibration factors. Ensure all parameters are within
the manufacturer's recommendations. If minor repairs are needed, conduct acceptance testing
and rerun the preliminary points. If major repairs are needed or the internal calibration factors
are adjusted, then the candidate transfer standard must undergo a new verification according to
Section 5.1.
© - Run one (1) cycle consisting of a minimum of six (6) concentration points (additional
concentration points are preferred) and a zero.
® - After the cycle is completed, review the data and verify they meet all requirements in Table
4-1. All equations, an example verification dataset and calculations can be found in Appendix A.
If any of the requirements are not met, a new verification must be completed. Assess the
problem and begin the verification process according to Section 5.1.
® - Ensure all documentation is complete and all records are saved to the appropriate data
storage system. The MO may place a summary report on the verified transfer standard which
states at a minimum: the date of the verification, the date of verification expiration, name of
person conducting the verification, make/model/SN, current internal calibration factors,
dates/slopes/intercepts of the original verification cycles, the average slope and the average
intercept. If Equation 10 in Appendix A is used in the traceability scheme, then the user should
also state the current equation to be used for calculating the standard concentration when
subsequently used. A summary report should be placed on the top of the transfer standard.
6.0 Operational Considerations of O3 Transfer Standards
6.1 How to Troubleshoot a Verification, Calibration, or Measurement Quality Check
Exceeding an Action Limit or Acceptance Criteria
When using a verified transfer standard to conduct a verification, calibration or measurement
quality check and the check is exceeding an action limit or the stated acceptance criteria it is not
acceptable to automatically assume the transfer standard is correct and conduct a calibration of
the O3 analyzer. The user should implement a systematic approach to determine what device is
causing the failure and further investigate to determine what component of the device is causing
the failure. When a transfer standard is found to be causing a failure, it must be assessed,
repaired and verified.
27
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Figure 6-1 provides a suggested approach to assess a failing check. This same approach can be
used for verifications, and reverification as well as calibration, and other measurement quality
checks of O3 analyzers.
Troubleshooting
Procedure
r
Conduct
Cal bra
Quiliiv
tian or
C 1 e: ¦-
PASS _
1
jsirg TRANSFER
STANDARD#1
Restrre Quality
Chszk l s ir g"
TRANSFER
STANDARD *1
Figure 6-1 Troubleshooting Flowchart
® - The user sets up and properly warms up verified transfer standard #1 at the site. The user
allows the indicated concentrations to stabilize. If the percent difference is within the action
limit or acceptance criterion, for the procedure the user should resume the check.
28
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
© - If the results exceed the action limits, the user should recheck all connections, fittings and
lines. If a failure is found, the user should correct the problem and resume the test.
© - If no failure is found, the user should recheck all other components such as the zero-air
system, and/or desiccant, etc. If a failure is found, the user should correct the problem and
resume the test.
© - If no failure is found, the user should retest the analyzer with an independent verified
transfer standard (transfer standard #2).
© - If the concentrations do not exceed the action limits or acceptance criteria the user should
resume the test with transfer standard #2. The user must ensure this transfer standard is
appropriate for the type of work being conducted. For example, the transfer standard used for
the APE must not be used for the calibration or routine QC checks. Transfer standard #1 must be
taken out of service for assessment, repair and verification.
© - If transfer standard #2 shows the same results as transfer standard #1, then the user should
resume the measurement quality check.
6.2 Additional Checks to Guard Against Systemic Data Loss
Users are cautioned that drift can occur in O3 transfer standards and calibrating an O3 monitor
with a transfer standard that has unknowingly drifted can cause systemic data loss in a
monitoring network. EPA recommends implementing a system of checks for early identification
of drift. All available information should be reviewed at least daily to detect measurement
system drift. The following are examples of additional checks that should be considered.
Establishing Action Limits - An action limit is a percentage of the minimum and maximum
values of the defined acceptance criteria. Corrective measures should be taken as soon as values
less than the lower action limit and higher than the upper action limit are exceeded. Action
limits can initially be set to approximately 50% of the acceptance criteria, however, this may
vary depending on the stability of the performance of the parameter being monitored or other
factors. Action limits should ultimately be set statistically based on historical data. One
approach is to set statistical action limits at plus or minus two standard deviations of the mean, or
at approximately 95% certainty. Where statistical action limits are outside of acceptance criteria,
corrective actions may be required.
EPA recommends that action limits be set and documented in QAPPs and SOPs. Setting action
limits and making the needed correction(s) to the measurement system when action limits are
exceeded will minimize the occurrences of exceeding acceptance criteria in transfer standards.
Conduct More Frequent Reverifications - Table 4-1 lists the minimum reverification
requirements. EPA strongly recommends that additional verifications or spot checks be
conducted.
Conduct Additional Checks of Transfer Standards at Monitoring Sites - There are many
simple ways to conduct additional checks of transfer standards. A simple check can be as easy
29
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
as quickly introducing a zero/span atmosphere into the transfer standard to assess the system. If
a MO is employing traceability Scheme #1 (Section 2.8.1 - using level 3 bench standards at
monitoring sites), additional reverifications can easily be conducted of the transfer standard at
the monitoring site using an independent transfer standard. For example, if an APE is being
conducted at an O3 monitoring site, the independent transfer standard used for the APE could
supply a sample to both the O3 analyzer and the Level 3 bench standard simultaneously.
"Buddy Site" Analysis - Buddy sites are O3 monitoring sites that are relatively close to each
other, known to have similar hourly concentrations and known to behave similarly under
differing environmental conditions. The buddy site check is a subjective test and must never be
used to judge data validity. This check is strictly a high-level visual review of a monitoring
system's functionality, including the transfer standards used for calibrations. For example, if two
monitoring sites are known to have similar peak concentrations or are known to rise and fall at
the same rate and at the same time each day, time series graphs could easily be made of the
buddy sites. The graphs could be transposed on one another and used to make a visual
determination about whether the buddy sites O3 analyzers are operating properly. If the graphs
do not line up as expected, then an investigation should occur immediately.
Many data logging systems can be programmed to conduct this analysis automatically. The user
can set buddy site parameters within the data logger and when the specified parameters are
exceeded the data logger can notify the user via email or even text messaging.
Review Data from Nightly Checks - These data can be used to determine if the measuring
system (including the transfer standards) is operating properly or if additional investigation of
the measurement system is warranted. Users can program data loggers or other controls to
automatically run nightly O3 checks. These checks may be the "official" QC check conducted
with a verified O3 transfer standard or may be conducted using the integrated zero span option
within the O3 analyzer. As stated in Section 3.2, the only appropriate use of the integrated
zero/span option within the O3 analyzer is to conduct automated zero/span checks (usually at
midnight each night). This data should only be used as a diagnostic check to inform the user
whether an action limit is breached, thereby triggering an investigation and possible recalibration
of the monitor prior to the point of data loss.
The user should set up the nightly check to run automatically and graph the results in a control
chart. The user should review the data and the general trend of the percent differences daily.
Once an action limit is reached, the user should conduct an assessment and take corrective
action. It is also possible to use a data logger or other system to automatically make this
determination and send warnings to alert the MO of a breach in action limits.
Review APE and Thru-the-Probe (TTP) Results - APE and TTP audits are conducted using
independent transfer standards. TTP audits are conducted using Level 2 field standards verified
quarterly. This provides a unique opportunity for the user to see if the overall traceability
scheme is functioning within the limits of this TAD. While an independent audit may show
failing results for many reasons, it is a good indicator of the overall accuracy of the O3 analyzer
and the transfer standard used for calibration. If the audit results are well within the action limit
30
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
or acceptance criteria, then this indicates that the overall monitoring system (including
traceability scheme) is functioning as designed. Results should also be compared to the other
measurement quality checks conducted at the site to assess whether bias exists with the transfer
standard used for calibration.
Conduct Intercomparison with Neighboring MO - Transfer standards can be used
conveniently to inter-compare among various agencies to assure accuracy and confidence within
the traceability scheme. For example, a MO could conduct 'across state line' comparisons in
joining NAAQS nonattainment areas to assess the accuracy of measurements. This could be
informative in areas where different legs of the traceability chain are adjacent geographically but
share a nonattainment area. For example, if one State MO conducts annual Level 2 verifications
with EPA Region 5 under SRP SN 6 and an adjacent State MO conducts annual Level 2
verifications under a different SRP, both organizations are equidistant from Level 1 and are
under a different Level 1. A successful intercomparison adds reassurances to each MO that
measurements were being conducted correctly and that their existing traceability schemes are
functioning as designed.
6.3 Determining the Calibration Scale and the Verification Points
Calibration Scale - The first step in determining what concentration points to run during a
verification is to determine a calibration scale that is low enough to encompass all possible
measurements expected and not so high that most of the measurements are on the low end of the
calibration scale. Generally, photometers in most O3 transfer standards are calibrated by
adjusting the zero and span. In the past, O3 monitors used for regulatory ambient air monitoring
were set to an upper range limit (URL) of 500 ppb. The calibration scale was based on the
selected URL and span adjustments were made at 80% of the URL (typically 400 ppb). Modern
FEM O3 photometers allow for more flexibility when selecting a calibration scale and the typical
high O3 concentrations measured are many times below 400 ppb. Therefore, a calibration scale
should be selected that covers the expected measured values and the NAAQS for O3, plus
additional room for potential anomalus high ambient O3 concentrations. The calibration scale
concept is fully described in the QA Handbook. The following procedure should be used to
determine a calibration scale:
1. Take the previous 3 years of 1-hour values. Determine the highest value.
2. Multiply the highest value by 1.5 to establish the calibration scale. If this value is below
the NAAQS, use 1.5 times the controlling NAAQS.
For example, if the 3-year high concentration (1-minute average values are preferred) within a
MO is 120 ppb, 1-minute measurements greater than 180 ppb (180 ppb is 1.5 times the highest
value) are not likely; therefore, a calibration scale of 180 ppb should safely encompass all
potential ambient O3 measurements.
31
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Level 1
(SRP)
Verification Points - A transfer standard must be verified over the concentration range it will be
used with the required number of concentration points (a minimum of 6 concentration points plus
a zero) approximately evenly spaced between zero and the selected calibration scale. Additional
points may be used (e.g., a point at the NAAQS). Although it is not a requirement to precisely
evenly space points, they must not be closely grouped together. For example, it would not be
acceptable to assay verification points of 200, 180, 170, 160, 150, 15, 0. It would be acceptable
to assay points evenly
spaced such as 200, 163,
126, 89, 52, 15, 0.
Users could also add a
70 ppb point to represent
the NAAQS
concentration or other
concentrations. Any
concentration point
assayed by a verified
transfer standard must
be equal to or less than
the highest point used
during its verification.
Therefore, it is
imperative to choose an
appropriate calibration
scale for the O3
monitoring network
prior to conducting
verification work since
this will determine the high point in any subsequent O3 monitor calibration.
Level 2
Highest verification point should be
slightly above the highest L3 point.
Level 3
Highest verification point should be slightly above the
calibration scale.
OB Monitor Calibrated at Monitoring Site
Example Calibration scale = 180 ppb
Figure 6-2 Example Verification Points for a Calibration Scale of 180 ppb
For example, Figure 6-2
shows suggested points
for a monitor calibration scale of 180 ppb. The highest point run during the Level 2 transfer
standard verification could then be 200 ppb and the highest point run during the Level 3 transfer
standard verification could then be 195 ppb. The Level 3 transfer standard would then have a
verification range of 0-195 ppb and must not be used at concentrations greater than 195 ppb.
6.4 Warm-up Time and Stability of Indicated Concentration
Transfer standards must be properly warmed up and be able to provide a stable output prior to
use. Operators should consult the manufacturer's instrument manual for specific
recommendations. Warm-up times will vary depending on how or if the transfer standard was
transported. Temperature plays a key role in how long it takes for a transfer standard to warm-up
and stabilize. Colder temperatures will have longer warm-up times. A minimum warm-up and
stabilization time should be no less than 30 minutes but could be much higher (i.e., overnight).
32
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
The user should set minimum criteria for warm-up times and objective criteria for output
stability and document these in the QAPP and SOP. At a minimum, EPA recommends that an
operator wait 5 additional minutes after the analyzer has begun to measure consistent,
instantaneous concentrations that show minimal variability and no discernible slope. In general,
the longer the operator waits to take a reading, the better the results. Additional information may
be found in the QA Handbook.
6.5 Timing and Frequency of Verifications and Reverifications
Significant planning must be considered when determining the verification and reverification
schedules. MOs should always have verified transfer standards at Levels 2 and/or 3, and
therefore MOs should map out an ideal annual schedule to follow each year for when the Level
2s are verified and subsequently when the Level 3s are verified. For areas where O3 monitors are
operating seasonally (i.e., March to October) Level 2 transfer standards should be completed by
January to give plenty of time for Level 3 transfer standards to be verified prior to conducting
calibrations at monitoring sites. For areas in which O3 monitors operate year round, it is
important to ensure verified transfer standards are continuously available at Level 2 and 3. EPA
recommends scheduling the Level 2 verification well in advance so that the service may be
completed without delay. EPA recommends that a tracking sheet or other mechanism be
employed to plan and conduct verifications so that verifications do not lapse. Under no
circumstances can a O3 transfer standard be used after the verification has lapsed; however,
Section 2.7 does allow for flexibility in conducting reverifications on transfer standards that have
lapsed. The traceability process must be fully documented in the QAPP and SOP.
6.6 Protecting Transfer Standards Against Damage
Transfer standards are highly sensitive electronic devices and depending on their designation
(bench versus field) may be frequently moved from one location to another location. Extreme
care must be taken to minimize damages to the devices, especially in the case of devices that are
regularly transported to field locations and used in routine air monitoring. Users should use great
care when placing devices on tables or lab benches by gently placing the devices and not setting
them down in a rough or carefree manner.
Transport can subject transfer standards to shock, vibration, temperature extremes and static
charge which can cause measurement drift. To minimize these conditions, transfer standards
should be properly padded while in transport and extreme temperatures should be avoided.
Proper padding may also be added to the inside of the chassis to ensure electric cards remain
seated. Care should also be taken to protect the transfer standards from static charges. Always
place a reminder sticker on the device if padding is used inside the chassis so that the user
remembers to remove this prior to turning the instrument on. Proper placement in the vehicle
and care when placing the equipment is also necessary to protect the integrity of the transfer
standard. Equipment should be strapped down if it is at risk for sliding around in the vehicle.
Furthermore, EPA recommends that padded cases be used for transport and that transfer
standards not be left in a vehicle in the hot sun or cold temperatures. EPA recommends that if
33
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
shipping is required, a specialized padded case or the original box and packing material (in good
condition) be used and that checks be conducted prior to and after shipping.
Every transfer standard must be traceable to a Level 2 bench standard. A Level 2 bench standard
is maintained at a fixed location except when taken to be verified against an SRP. When kept
under laboratory conditions, the integrity of the Level 2 bench standard can be preserved. Under
the controlled conditions at the fixed location, variability in the Level 2 bench standard will be
reduced, providing better accuracy and uniformity among all O3 analyzers in the network. If
doubt arises with the proper functionality of a transfer standard (e.g. due to rough treatment in
the field) it can be brought back in for reverification to the Level 2 bench standard.
6.7 Maintaining Backup Verified Transfer Standards
EPA recommends that MOs always have access to back up transfer standards at each level. This
is especially true for Level 2s. A Level 2 transfer standard must be transported to one of only 10
locations in the US to be verified against a Level 1 (SRP); this can be very time consuming and
is dependent upon the availability of the SRP operator and the MO travel budget. If a Level 2
transfer standard unexpectedly fails in the middle of O3 season, it may be impossible to repair
and verify it before it is needed. This can result in the MO not having the needed Level 3
transfer standards necessary to seamlessly conduct calibrations or other quality work at their
monitoring sites.
6.8 Preventive Maintenance
Preventive maintenance must be conducted consistently to be an effective way to minimize
malfunctions and downtime. All commercially available devices have manuals that describe
preventive maintenance items and frequencies. At a minimum the as-found and as-left
conditions of all diagnostics should be observed and documented for all preventive maintenance
activities. Assaying ozone test concentrations can also be helpful as an as-found test especially
for transfer standards exhibiting abnormal performance. Acceptance testing should also be
conducted following preventive maintenance. EPA recommends that transfer standard owners
complete and document all preventive maintenance according to the instrument manual as part of
and prior to normal verifications or reverifications. Laboratories providing O3 transfer standard
verification services should review documentation showing that preventive maintenance has
been completed. Verification services may be refused if this documentation is not presented and
current.
6.9 Transfer Standard Repairs
O3 transfer standards will likely experience damage or malfunction requiring assessment, repair
and verification. Major repairs are likely to require the device to be verified. Minor repairs or
preventive maintenance are not likely to require the device to be verified. The MO should
always follow the manufacturers' recommendations when making repairs. Acceptance testing
must always be conducted following any repair.
34
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
An as-found test and as-left test should also be conducted to objectify what action to take
following a repair (e.g., take no action, conduct verification or conduct reverification). This
process should be clearly defined in the QAPP and SOP and a standard form should be used to
document what happened. In general, if readings return to normal after acceptance testing, the
instrument verification may still be valid. In any case, EPA strongly recommends that a MO
have a well-defined and documented approach to repairs and verification. The following table
lists some examples of major and minor repairs.
Table 6-1 List of Common Major and Minor Repairs for O3 Transfer Standards
Major Repair (requires verification)
Minor Repair
Detector or Optical Bench
malfunction/replacement
Preventive Maintenance as described in
the instrument manual
Optical Bench Lamp Replacement
Pump Replacement, rebuilding or other
flow related repairs
Photometer tube replacement
Any O3 generator component
Pressure sensor replacement or
recalibration
UV Bench Lamp Adjustment
Temperature sensor replacement or
recalibration
Display, Keypad or related boards
Detector or related board replacement
Minor Leak Repair (i.e., tighten a fitting
or worn teflon line)
Major Leak repair (i.e., reseat a solenoid
valve)
Fuse Replacement
Motherboard Replacement
Fan Replacement
Optical Bench Temperature Calibration
All input/output board replacement
6.10 Spare Parts Inventory
The purpose of a spare parts inventory is to minimize downtime when equipment failures occur.
For organizations that own many transfer standards EPA recommends that spare parts be on hand
for needed repairs. It is not uncommon for transfer standards to require troubleshooting and
repair. A listing of spare parts is always available in the equipment manual. Spare parts may be
purchased as part of a larger equipment procurement. Users can develop an inventory based on
anticipated needs for the spare parts. Preventive maintenance items such as pump rebuild kits,
UV lamps, etc. should always be on hand for periodic maintenance or repairs. Other items such
as electronics boards, detectors, power supplies, etc. should be available, but may be needed less
frequently. Therefore, the user should have less of these on hand or at a minimum have a
method to receive these parts quickly. An appropriate spare parts inventory should be kept up-
to-date and replenished when parts are used.
35
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
6.11 Zero-air Supply
Contamination, humidity, pressure or other problems with zero-air can be a common source of
measurement error in O3 transfer standards. 40 CFR Part 50 Appendix D § 4.4.1 defines zero-air
as follows: "The zero-air must be free of contaminants which would cause a detectable response
from the O3 analyzer, and it should be free of NO, C2H4, and other species which react with O3".
Within a transfer standard, zero-air is pushed through an O3 generator to generate a test
atmosphere. Zero-air is also used to provide a reference measurement for the assay of the
sample. The same zero-air source must be used for each component and for both the candidate
transfer standard and the standard of higher authority during a verification. Many configurations
are acceptable; however, users are cautioned that zero-air must be introduced to the O3 transfer
standard according to the manufacturer's recommendation using a method that does not alter the
zero-air or the pressures within the UV measurement cycles. EPA recommends viewing the
photometer sample pressure during the measurement cycle to determine if a change in pressure is
occurring when the valve alternates between the reference and sample cycle.
Many adequate zero-air systems are commercially available and should be utilized. Users should
utilize oil free air compressors in zero-air systems. It is also acceptable to build a zero-air system
using a pump, pressure regulator, silica gel, activated charcoal and molecular sieves. In either
case, the zero-air systems must be properly maintained and tested in accordance with the
manufacturer's specifications and/or the MO's QAPP or SOP. All maintenance activities must be
documented. The QA Handbook provides additional information on adequate zero-air and zero-
air testing.
6.12 Excess Flow
There are different ways to set up a flowing system to conduct verifications of O3 transfer
standards. A typical O3 transfer standard system consists of an O3 generation device supplied
with metered zero-air, a manifold to allow for multiple devices to measure the generator output,
and a verified photometer to measure the generated O3. Many of these devices are preassembled
and commercially available. Each commercially available device will have specifications for
zero-air pressure, photometer flow and excess flow in the instrument manual. Operators must
follow these specifications as they are the design criteria for the instrument. For all devices, the
O3 generator must have a stable flow and provide enough flow to supply both the candidate
transfer standard and the verified transfer standard so they do not pull in room air. Users must be
aware of the total output flow of the generator and total photometer flows in both instruments.
6.13 Protecting Internal Components of Transfer Standards
Transfer standards should never be allowed to sample room air unfiltered from room dust or
other contaminants. It is best practice to only allow a transfer standard to pull air from a suitable
source of zero-air. A particulate filter may be added to the inlet of the transfer standard or the
photometer pump can be turned off to protect internal components.
36
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
6.14 Acceptable Materials
Borosilicate glass, FEP Teflon® or their equivalent must be the only material in the sampling
train that can be in contact with an ozonated sample while using an O3 transfer standard. Other
materials such as stainless steel or brass may be used in the zero-air system or with any
downstream component of the measurement system.
6.15 Line Conditioning and Line Length
O3 is a highly reactive gas. An atmosphere of O3 will be greatly impacted by an improperly
conditioned or dirty measurement system. All teflon lines and surfaces in contact with an
ozonated sample can scavenge O3 especially in the presence of dust or debris and must be
properly conditioned prior to use. The user should run an ozonated sample (i.e., run the highest
O3 concentration in the traceability scheme) through the transfer standard and any lines used to
adequately condition the system. After conditioning, the transfer standard should be run
ensuring that all components (lines, fittings, etc.) that will be in contact with the ozonated sample
are in line. In other words, the conditioning and testing should be conducted using the same
configurations as is used during verifications, calibrations or other measurement quality checks.
If O3 scavenging is still occuring, then the lines should be reconditioned or replaced as
necessary.
The measurement system being conditioned should be vented to an exhaust hood or charcoal
column so that high concentrations of O3 are not being exhausted into the laboratory and inhaled
by the operator. The length of line used during a verification will have a direct impact on the
time it takes each O3 point to stabilize during a verification and can also impact the stability of
the O3 readings. The length of teflon line used to deliver test atmospheres to candidate transfer
standards should be kept to a minimum (i.e., less than 1 meter if possible). EPA recommends
that the same set of lines be used for all verifications. The lines should be capped, labelled and
kept free from dust or other debris or chemicals when not in use. Operators should have a few
sets of lines which are properly conditioned, labelled and stored in a dedicated location. Having
multiple sets of known good lines can also aide in troubleshooting.
6.16 Acceptable Testing Devices
All testing devices used in the acceptance testing and qualification testing must be NIST
traceable. These devices include at least temperature, pressure and digital multimeters. Each
device should be recertified annually or according to other guidance and be clearly labelled
stating, at a minimum, the certification expiration date and entity who conducted the
certification. A certification certificate must be filed and stored according to data storage and
recordkeeping procedures. EPA recommends that 2 devices be available for each measurement
needed and that these devices be certified on alternating schedules so that the user can have a
secondary spot check of the measurement, (see GMP 13-2018 Section 1.1.1 NOTE 4)
37
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
6.17 Data Handling and Documentation
It is paramount to the transfer standards traceability that a data handling process be in place. All
data relevant to the qualification, acceptance testing, verification and reverification must be
organized so that information can be accessed in the future. Each transfer standard should have a
file which includes all historical information for the device. A file naming convention and file
folder structure should be maintained and documented. A summary sheet should be developed
and affixed to the transfer standard so that the user can easily see that the device is within
verification limits and see when the current verification expires. This will not be the official
verification report but will be used to quickly reference the verification summary.
This process must be documented according to a formal records management policy. The QA
Handbook provides further information on records management.
6.18 Use of Standard Forms and Logbooks
Developing a consistent technique for documenting information in a logbook and archiving this
information is very important. Standard forms should be developed which allow for consistent
and thorough documentation of the conditions at the time of use of the transfer standard and the
results of the work completed. Standardized forms for maintenance/repairs, acceptance testing,
verifications, calibrations and other routine work are needed to provide this documentation. A
proper paper or electronic logbook may also be used. The QA Handbook provides more
information about logbooks and documentation.
6.19 Work Area Cleanliness and Organization
Good laboratory practices dictate that work areas should be kept clean and organized and it is
important to maintain a dust and contaminant-free environment in the laboratory or air
monitoring sites where transfer standards are being used. Benches and other surfaces should be
wiped down frequently to minimize the potential for contamination of the photometer cell or
other critical components in the device. All candidate transfer standards should be checked for
excessive dust or other problems inside the chassis of the instrument. Users should keep in mind
that the O3 measurement principle (photometry) is based on absorption of UV light and that O3 is
a highly reactive molecule. Even the smallest amount of dust or other contaminant (Hg,
hydrocarbons, bug sprays, perfumes, etc.) will cause large variations in measurements.
Laboratory cleaning and organizing should be part of a regular preventive maintenance program.
Labs that handle large numbers of transfer standards should have an area designated for
receiving, an area designated for transfer standards in the verification process, and an area where
transfer standards are completed and ready to pick up. Signage should be placed to clearly
delineate one area from another.
38
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
6.20 Laboratory Conditions (Temperature, Relative Humidity)
All transfer standards must be operated within the operating temperature range specified in the
instrument manual. Swings in temperature or humidity will impact the operation of transfer
standards; therefore, the verification laboratory conditions should be kept relatively constant to
within approximately 20-30 degrees C. A certified temperature and relative humidity device can
be installed and observed by the user to ensure these conditions are met. Most laboratory air
handling systems will control the conditions to within acceptable limits provided the systems are
not malfunctioning. In locations where more extreme temperature or humidity shifts are
possible, it may be necessary to have additional air handling systems in place and the operator
may need to take additional actions to ensure the laboratory conditions are acceptable. EPA
recommends that the user document the current lab conditions or otherwise record the laboratory
conditions continuously as part of the verification.
6.21 Use of Data Loggers or other Automation Software to Run Verifications
When conducting O3 verifications it is necessary to generate and assay different concentration
points and record the concentrations of the candidate transfer standard and the transfer standard
of higher authority. In past years, this required an operator to be in front of these devices while a
verification was occurring and manually write down data and change settings to complete a
verification. Many different options are available today. MOs can easily set up calibration
routines in data loggers or other laboratory software systems to automatically run through the
concentration points and record the resultant data. Simple computer programs can also be
written to automatically control the O3 generator and record the data. In either case, the MO will
need to have a written procedure and pay close attention to the resultant data. The advantage to
the automation is that a person can go do other work while O3 transfer standards are being
verified and multiple verifications can be conducted simultaneously. After the verification cycle
completes, the operator can review the data and set up a new verification. Verifications could
also be started at the end of a day and allowed to run overnight.
6.22 Tools
A quality tool set is required to conduct O3 verification work. The most commonly needed tools
are a 9/16-inch wrench (a full set is recommended), adjustable pliers or channel locks, line
cutters or wire cutters, LED flashlight, magnifying glass, and utility knife. However, many other
tools may be needed to conduct this work. A full laboratory tool set is recommended including a
vise, soldering equipment, full drill set, and other tools needed for unforeseen repairs or for
completing electrical or pneumatic connections.
6.23 Safety
The user must be aware that they are working with electricity and a harmful gas (O3). Therefore,
any organization conducting this work must have adequate safety training in place. The user
39
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
must also read all manufacturers' safety warnings and follow all laboratory safety requirements
(i.e., properly vented exhaust lines, safety glasses, etc).
40
-------
Transfer Standards for the Calibration of Air Monitoring Analyzers for Ozone,
January 2023
Additional References
Norris, J.E., Band, A.H., Biss, R.J., Guenther, F.R., 2004. Upgrade and intercomparison of the
U.S. Environmental Protection agencies ozone reference standards, in: Proceedings of the
Air and Waste Management Association's Annual Meeting and Exhibition.
Norris, J.E., Choquette, S.J., Viallon, J., Moussay, P., Wielgosz, R., Guenther, F.R., 2013.
Temperature measurement and optical path-length bias improvement modifications to
National Institute of Standards and Technology ozone reference standards. Journal of the
Air and Waste Management Association, https://doi.org/10.1080/10962247.2013.773951
Paur, R.J., A.M. Bass, J.E. Norris, T.J. Buckley, 2003. Standard Reference Photometer for the
Assay of Ozone in Calibration Atmospheres.
Thompson, A., Taylor, B., 2008. Guide for the International System of Units (SI).
Viallon, J., Moussay, P., Norris, J.E., Guenther, F.R., Wielgosz, R.I., 2006. A study of
systematic biases and measurement uncertainties in ozone mole fraction measurements with
the NIST Standard Reference Photometer. Metrologia 43, 441-450.
41
-------
Appendix A Equations and Example Calculations
This appendix shows an example verification and reverification for a Level 3 transfer standard
("candidate standard"). Verification example data and calculations are presented first, followed
by reverification. EPA recommends using a spreadsheet or other automated data entry system
containing embedded equations that perform the calculations automatically. EPA also
recommends that all users of this TAD understand how to manually conduct the calculations.
All verifications of transfer standards must be completed according to Section 5.0 and Table
4-1. A verification consists of three (3) testing cycles (performed independently) in which the
candidate standard's measurements are compared to those of the bench standard (i.e.,, the
"standard of higher authority"). Table A-l provides an example of verification data for a Level 3
candidate transfer standard used in a monitoring network having a calibration scale of 180 ppb,
along with the corresponding data from the Level 2 bench standard. For both standards,
measurements were taken at each of eight (8) concentrations ("points") distributed along the
workable range of concentrations.
Table A-l. Example Verification Data (ppb)
Cvclc 1 (/=!)
Cvclc
2 (i=2)
Cvclc 3 (i=3)
Level 3
Level 3
Level 3
Cone.
Level 2
Ciiniliihilc
l.c\cl 2
Ciiiulidiilc
Level 2
(:iniliil:itc
Tcsl
lie iicli
Tninsl'cr
licncli
Tninsl'cr
Bench
Ti'iinsl'ci'
Point
Sliinihi ril
Sin ml :i ril
St 2iiicl:i ril
St ;i ihI ii ril
Sl2inil;iril
S121111I21 ril
(/')
(•V/,:)
(r/;)
(A.,)
(.!"-/)
(¦Vv)
(.¦¦«;)
1
0.2
-0.1
0.2
0.0
0.0
0.3
2
200.0
200.9
199.5
201.1
200.1
201.4
3
163.0
163.7
162.6
163.9
163.3
164.3
4
126.0
126.5
126.4
127.3
126.2
127.0
5
89.0
89.3
88.9
89.5
89.5
90.0
6
70.0
70.3
69.9
70.3
70.3
70.7
7
52.0
52.2
51.9
52.2
52.2
52.5
8
15.0
15.0
14.9
14.8
15.0
15.0
Note: Bench standard = standard of higher authority.
The subsections which follow reference a series of equations for statistics that assess the success
or failure of a given verification or reverification test, along with the acceptance criteria used to
determine success based on each statistic. Each calculation is illustrated using the above
example data from Table A-l.
A-l
-------
Al. Per Point Percent Difference (% Diff) in Measured Concentration
For each test point within a cycle corresponding to a concentration above 50 ppb (i.e.. for
7=2 through 7 in Table A-l), the percent difference (% Diff) in measurements between the
candidate transfer standard and the standard of higher authority is calculated using Equation 1:
Equation 1 - Percent difference (% Diff) in measured concentration at the jth
concentration point within the ith cycle
%Difftj = {{ytj - xtj) h- xtj] x 100%
where: (xij) = Measured concentration from the standard of higher authority.
{yij) = Measured concentration from the candidate transfer standard.
At each concentration test point above 50 ppb, the calculated value of %oDiffj must be within
±3.1% to be considered acceptable. If this criterion fails at any of these test points, the test
cycle's data are disregarded and the cycle is repeated.
To illustrate this calculation, consider the reported data at the highest concentration point in
cycle #1 (i.e.,, i=l,j=2) within Table A-l. Here,
%Diff12 = {(200.9 - 200.0) - 200.0} X 100%
= 0.45%.
This value falls within the acceptance criterion of ±3.1%.
A2. Per Point Absolute Difference (AbsDiff) in Measured Concentration
For each test point within a cycle corresponding to a concentration at or below 50 ppb (i.e.,
for7=1 and7=8 in Table A-l), the absolute difference (AbsDiff) in measurements between the
candidate transfer standard and the standard of higher authority is calculated using Equation 2.
Using the same notation as for Equation 1 above:
Equation 2 - Absolute difference (,AbsDiff) in measured concentration at the fh
concentration point within the ith cycle
AbsDiff jj = yi} - xi}
At each test point at or below 50 ppb, the calculated value of AbsDiffj must be within ±1.5 ppb
to be considered acceptable. If this criterion fails at any of these test points, the test cycle's data
are disregarded and the cycle is repeated.
A-2
-------
To illustrate this calculation, consider the reported data at the lowest concentration point in cycle
2 (i.e., i=2,j=S) in Table A-l. Here,
Abs Diff28 = 14.8 - 14.9
= —0.1 ppb
This value falls within the acceptance criterion of ±1.5 ppb.
A3. Least Squares Linear Regression Analysis (Slope and Intercept Estimation)
For each cycle within a verification test (and for a reverification test cycle), an ordinary least
squares linear regression line is fitted to data from all concentration test points to predict the
measurement from the candidate transfer standard as a simple linear function of the measurement
from the standard of higher authority. If subscript i refers to the cycle (i = 1, 2, 3), the following
notation is used:
(»«) = Number of concentration test points in the 7th cycle with data reported (//;=8 in
Table A-l for each cycle).
(xij) = Measurement from the standard of higher authority for the/h concentration in
the 7th cycle (j= 1, ..., rii).
(yij) = Measurement from the candidate transfer standard for the/h concentration in
the 7th cycle (j= 1, ..., rii).
(£ij) = Absolute difference between yij and the value predicted by the regression
model (yij)., for the jth concentration in the 7th cycle (j= 1, ..., //,), also known as
"residuals."
(jHz) = Intercept of the linear regression line for the 7th cycle (needs to be estimated).
(fii) = Slope of the linear regression line for the 7th cycle (needs to be estimated).
(*i) = Arithmetic mean of values x,, (j= 1, ..., rii) within the 7th cycle (see Equations
5 and 6 for arithmetic mean formula).
= Arithmetic mean of values^ (j= 1, ..., //,) within the 7th cycle (see Equations
5 and 6 for arithmetic mean formula).
(nti) = The estimated intercept of the best fit line for the 7th cycle (using the cycle
measurements Xij and
(bi) = The estimated slope of the best fit line for the 7th cycle (using the cycle
measurements Xij and
Equation 3 - Linear regression model for the ith cycle (7 = 1,2,3)
Vij — Pi ^ f^i^-i) £ij (j l,-..,Wj)
A-3
-------
Equation 4 - Least squares estimates for slope and intercept for the ith cycle (z =
1,2,3)
Slope estimate (unitless):
m, =
Intercept estimate (ppb): bt = Kj — wijXj
Equation 5 - Predicted concentration reported by the candidate transfer standard
according to the fitted line within the ith cycle (z = 1,2,3)
yi] = bi + mixij
For a given test cycle, the acceptance criteria for the fitted regression line (Equations 4 and 5) are
as follows:
• The value of the slope estimate nti (Equation 4) must fall within 1.00 ± 0.03 (i.e.,, within
±3% of 1.00).
• The value of the intercept estimate bt (Equation 4) must fall within 0.00 ± 3 ppb.
If either of these two criteria fails for a given test cycle, the test cycle's data are disregarded and
the cycle is repeated.
Using the data from Table A-l, the fitted regression line for each of the three example test cycles
has slope and intercept estimates listed in Table A-2. Each of these estimates achieves its
respective acceptance criterion.
Table A-2. Slope and Intercept Estimates for Each Fitted Regression Line (one line per
cycle) in the Example Verification Data
Cycle (i)
Slope (nti)
Intercept (bt)
1
1.0053
-0.1518
2
1.0091
-0.2136
3
1.0058
0.0511
A4. Average slope (m)
Across the three cycles of a verification test for the candidate transfer standard, the arithmetic
average of the estimated slopes from the three cycles is calculated. Using notation from Section
A3, the average slope is calculated as follows:
Equation 6 - Average slope (m) for a verification test
A-4
-------
Using the slope estimates from Table A-2, the average slope for the example verification data is
calculated as
m = (1.0053 + 1.0091 + 1.0058) h- 3
= 1.0068.
A5. Average intercept (b)
Across the three cycles of a verification test for the candidate transfer standard, the arithmetic
average of the estimated intercepts from the three cycles is calculated. Using notation from
Section A3, the average intercept is calculated as follows:
Equation 7 - Average intercept (b) for a verification test
i=i
Using the intercept estimates from Table A-2,
b = {(-0.1518) + (-0.2136) + (0.0511)} h- 3
= -0.1048.
A6. Stability Estimate for the Slopes Within a Verification Test (Standard Deviation ~
SDm)
Across the three (3) cycles of a verification test for the candidate transfer standard, the
standard deviation of the estimated slopes from the three (3) fitted regression lines is calculated
as follows (using notation from Section A3):
Equation 8 - Standard deviation of the three slopes within a verification test (SDm)
The calculated value of SDm must be below 0.0075; otherwise, the entire verification test must be
repeated.
Using the slope estimates from Table A-2 and the value of m from Section A.4, the standard
deviation of the slopes calculated from the example verification data is calculated as
SDm = -y/[(1.0053 - 1.0068)2 + (1.0091 - 1.0068)2 + (1.0058 - 1.0068)2] -h 3
= ^/[(0.00000205 + 0.00000551 + 0.00000084] -h 3 = 0.00167
A-5
-------
This value falls below 0.0075 and thus achieves the acceptance criterion.
A7. Stability Estimate for the Intercepts Within a Verification Test (Standard Deviation
- SDb)
Across the three (3) cycles of a verification test for the candidate transfer standard, the
standard deviation of the estimated intercepts from the three (3) fitted regression lines is
calculated as follows (using notation from Section A3):
Equation 9 - Standard Deviation of the three intercepts within a verification test
(SDb)
sd>=]I(z L(b'-~b)2)
The calculated value of SDb must be below 1.00; otherwise, the entire verification test must be
repeated.
Using the intercept estimates from Table A-2 and the value of b from Section A.5, the standard
deviation of the intercepts calculated from the example verification data is calculated as
SDb = V[((—0.1518) - (—0.1048))2 + ((-0.2136) - (-0.1048))2 + (0.0511 - (-0.1048))2] -h 3
= V[(0-00221 + 0.01184 + 0.02429] -h 3 = 0.11306
This value falls below 1.00 and thus achieves the acceptance criterion.
A8. Standard Concentration for the Candidate Transfer Standard
For the candidate transfer standard, the method to determine the value of its standard
concentration at each concentration point within a test cycle depends upon how the unbroken
chain of calibrations is maintained for meeting NIST traceability (Section 2.1). Users must
determine and document their chosen method of maintaining the unbroken chain of calibrations.
The three options for the method to determine the standard concentration are as follows:
Option #1: The internal calibration factors for the transfer standard's photometer are
adjusted via a calibration which is based on the standard concentration of the
higher authority transfer standard.
Option #2: The internal calibration factors for the transfer standard's photometer are set to
one (for the span) and zero (for the zero), a least squares linear regression line
(Equation 3) is fitted to the data, and the slope and intercept estimates of this
fitted line (Equation 4) are noted. Equation 10 is then used to calculate the
standard concentration when the transfer standard is subsequently employed.
A-6
-------
Option #3: A combination of Options #1 and #2.
The selected method must be consistent throughout the traceability chain and with the
procedures in this TAD, and it must be well documented in the QAPP and SOP.
If either Option #2 or Option #3 is implemented, the standard concentration at a given
concentration point within the test cycle is calculated as follows:
Equation 10 - Standard Concentration (ppb) at a given concentration point
1
Standard O3 Cone = —(Indicated O3 Cone — b)
m
where: (m) = The slope of the test cycle's fitted least squares regression line, as determined by
Equation 4.
(b) = The intercept of the test cycle's fitted least squares regression line (ppb), as
determined by Equation 4.
(Indicated O3 Cone) = The concentration value (ppb) reported by the candidate transfer
standard at the given concentration point, either from its front panel, calibrated analog
voltage output, digital output, or some other data reporting mechanism.
To illustrate this calculation, consider the second concentration point within cycle #1 in the
verification example within Table A-1. Here, the candidate transfer standard reported a
concentration of 200.9 ppb. Then considering the slope and intercept estimates for this cycle
which are given in Table A-2, the standard concentration would be:
1
Standard OsConc = QQgg {200.9 — (—0.1518)}
= 200.0 ppb
A9. Reverification Testing and Acceptance Criteria
All reverification tests for a given transfer standard must be completed according to Section 5.2
and Table 4-1. A reverification test involves performing a single testing cycle. The linear
regression model specified in Equation 3 above is fitted to the data produced upon executing this
testing cycle.
Use the following steps to determine whether or not a reverification test is successful:
1. Use the data collected during the reverification test cycle to calculate the following:
a. Percent difference (% Biff) in measured concentration, calculated at each tested
concentration point above 50 ppb (Equation 1).
b. Absolute difference (AbsDiff) in measured concentration, calculated at each
tested concentration point at or below 50 ppb (Equation 2).
A-7
-------
c. Least squares estimates for the slope (m) and intercept (b) of the cycle's fitted
regression line (Equation 4).
Compare each of the calculated values in #1 against the following acceptance criteria to
determine if the acceptance criteria are achieved:
a. For a given testing concentration point above 50 ppb, the calculated value of
%Diff must be within ±3.1% to be considered acceptable.
b. For a given testing concentration point at or below 50 ppb, the calculated value of
AbsDiffij must be within ±1.5 ppb to be considered acceptable.
c. The calculated value of m must fall within ±0.015 of the candidate transfer
standard's calculated mean slope from the most recent successful verification test
(Equation 6) to be considered acceptable.
d. The calculated value of b must fall within ±1.5 ppb of the candidate transfer
standard's calculated mean intercept from the most recent successful verification
test (Equation 6) to be considered acceptable.
If all acceptance criteria are achieved for the given reverification test cycle (i.e., either
#2a or #2b is achieved across all concentration testing points, AND #2c and #2d are
achieved for the fitted regression line), then for the given candidate transfer standard,
a. Classify the reverification test as having successfully passed.
b. Using Equation 10 (and if either Option #2 or #3 above is used), use the updated
slope and intercept estimates (from #lc) to calculate the standard concentration
when the transfer standard is subsequently used.
If one or more instances are noted where acceptance criteria in #2 were not achieved in
the reverification test, then the candidate transfer standard should be assessed for
problems and, upon resolving any problems, must undergo and pass a full three-cycle
verification test (Section 5.1) before it can be used again.
A-8
-------
Appendix B Example Acceptance Testing Data Sheet19
Ozone Transfer Standard Acceptance Testing
1. Complete acceptance testing after proper warm up and while sampling zero air.
2. Compare all readings to the manufacturers recommendation's.
3. Readings not meeting the manufacturers recommendation must be corrected prior to conducting verification.
Operator:
Organization:
Instrument Make:
Instrument Model:
Instrument SN:
Date Preventive Maintenance Performed:
Transfer Standard Role:
PARAMETERS
Prior to Transport
As Found
As Left
DATF
TiMi:
I.AB TI \1PI RATCRI
I.AB STANDARD PRFSSIJRF
SIX >PF (CALIBRA TION FACTOR)
/Flic) (CAI.IBRATK >N FACT* >R)
SAMPI.F PRFSSURF
SAMl'FI I IMl'l RA I KRI
C I I.I 1 INTFNSITY
CFI.I. 2 INTFNSITY (if dual cell)
PIIC)K )MI 11 R FI,OW
PIIC >T( )MI 11 R LAMP Ti:MP
boxtfmp
-Addilioiuil l';ir;iniclcrs. Add IL-re-
-Addilioiuil l';ir;nnclcrs. Add IL-iv-
- Additional l'iiramda-s. Add I Icrc-
( OMMIA IS i milt1 nil 1il 111 |iiii'.iiiirlris Mini iIm mil lllci-l 1 lit- iiiiiiiiiriicliiix'is vpi-iilicilioil-i
19 This form is available at: https://www.epa.gov/aintic
B-l
-------
Appendix C Utilizing NIST 7 Essential Elements of
Traceability
NIST has developed a document entitled Good Measurement Practice for Ensuring Me trological
Traceability (GMP 13)20. The stated purpose of this document is to "enable compliance with
essential elements of Metrological Traceability". It is the transfer standard provider's
responsibility to develop procedures and other quality documentation to support their
claim of traceability and to ensure that the transfer standard is used appropriately. This
TAD provides specific procedures for transfer standard providers and users to follow to
meet the claim of traceability for their O3 measurements. EPA recommends that all
organizations become familiar with the concepts in GMP 13. EPA's O3 traceability scheme
meets the NIST GMP 13 guidance and "7 Essential Elements" as described below.
CI. Realization of SI Units
EPA follows the Guide for the Use of the International Systems of Units (SI)21. While part per
million (ppm) and part per billion (ppb) generally are not acceptable SI units, the SI does allow
for exceptions per the note in Section 7.10.3 of the SI guide. To avoid confusion among users of
this TAD and because the CFR uses ppm and ppb to express the value of quantity (rather than
nmol/mol), EPA uses ppm and ppb as the value of quantity for O3 concentrations in this TAD.
Thus, EPA regards ppm and ppb as acceptable units for use in all documentation of EPA's
traceability scheme (Thompson and Taylor, 2008).
C2. Unbroken Chain of Calibrations
EPA's O3 transfer standards transfer traceability from higher authority to lower authority by
comparing the NIST Level 1 SRP O3 standard to EPA Level 1 SRPs, Level 1 to Level 2, and
Level 2 to Level 3. Measurements from the NIST Level 1 SRP are linked to the BIPM (i.e., the
International Bureau of Weights and Measurements) through participation in the Protocol for the
Key Comparison BIPM.QM-K1, Ozone at ambient level22. The NIST Level 1 SRP is compared
to the BIPM SRP biannually. The NIST Level 1 SRP linked to the BIPM is considered the
highest authority standard in the United States. The EPA Office of Research and Development
(ORD) Level 1 SRP is compared to the NIST Level 1 SRP annually. EPA Regional Level 1
SRP's are compared to the EPA ORD Level 1 SRP annually. Level 2 transfer standards are
compared to the EPA Regional (or ORD) Level 1 SRP annually. Level 3 transfer standards are
compared to Level 2 transfer standards either annually or semiannually (depending on use). A
hierarchy diagram and detailed description of EPA's traceability scheme can be found in Section
2.0.
Beyond the Level 1 SRP, traceability is transferred by comparing the standard concentration of
the higher authority transfer standard versus the indicated concentration of the lower authority
transfer standard (candidate transfer standard) at various concentrations (at least 6 different
20 https://www.nist.gov/svstem/files/documents/2019/06/21/gmp-13-ensuring-traceabilitv-20190621.pdf
21 https://phvsics.nist. gov/cuu/pdf/sp8.1. .1. .pdf
22 https://www.bipm.org/ntils/en/pdf/BIPM.OM-Kl protocol.pdf
C-l
-------
concentrations and a zero are required). The unbroken chain of calibrations is established by one
of three options:
1) The candidate transfer standard photometer internal calibration factors are adjusted by
means of a calibration based on the standard concentration of the higher authority transfer
standard;
2) The candidate transfer standard internal calibration factors are set to one (for the span)
and zero (for the zero), a least squares linear regression is calculated, and Equation 10 of
Appendix A is used to calculate the standard concentration when the transfer standard is
subsequently employed; or
3) A combination of options 1 and 2.
The method used should be consistent throughout the traceability chain and must be consistent
with the procedures in this TAD and be well documented in the QAPP and SOP.
C3. Documented Calibration Program
All work encompassing traceability of O3 measurements must be documented in a QAPP and
SOP. Each step in the EPA O3 traceability scheme is completed at defined frequencies. Table
4-1 of this document outlines the required verification intervals and acceptance criteria for each
level of transfer standards. Other devices used in EPA's traceability scheme (i.e., pressure,
temperature, and voltmeter) are also required to be verified on a regular basis. These devices and
frequencies are stated in this document.
C4. Documented Measurement Uncertainty
All Level 1 SRPs are designed and built by NIST to meet the specifications in Standard
Reference Photometer for the Assay of Ozone in Calibration Atmospheres, NI STIR 6369.23
(Paur et al., 2003) The design specification includes an uncertainty budget for the measurement
components. All SRPs used within the EPA's O3 traceability scheme are designed and run
identically and therefore the uncertainty is considered to be identical throughout the SRP
network. NIST has conducted subsequent studies to test the design specifications and has made
several improvements throughout EPA's SRP network which were also rigorously tested and
documented. (N orris et al., 2013, 2004; Vial Ion et al., 2006)
Implementing the O3 Uncertainty Budget into the SRP Network
1. NIST Level 1 SRP comparison to ORD Level 1 SRP - NIST provides a report stating the
estimated expanded uncertainties of the SRP O3 concentration defined by the following
equation:
u{x) = V(0.28)2 + (1.1 X 10-2x)2 nmol/mol
23 https://www.nlst.gov/piiblications/standard-reference-photoineter-assay-ozone-calibration-atmospheres
C-2
-------
2. ORD Level 1 SRP comparison to EPA Level 1 SRP - ORD incorporates the uncertainty
budget received from NIST into their Verification Reports for the SRP comparisons. For
each of the SRPs, the report includes the Expanded Uncertainty. The report includes the
uncertainty at the highest concentration, the lowest concentration and the uncertainty at
70 nm/mol (ppbv). The report will also include the formula to be used in calculating the
expanded uncertainty.
C5. Documented Measurement Procedure
EPA has written procedures and guidance for all aspects of O3 monitoring and requires transfer
standard users conducting regulatory O3 monitoring to have proper written QAPPs, SOPs,
validation methods and follow good laboratory practices and equipment handling for providing
verification results with accurate and traceable values with appropriate uncertainties.
A certificate and report must be provided to the transfer standard owner after verification
detailing the verification results. Users should review the NIST SOP l24, Recommended
Standard Operating Procedure for Calibration Certificate Program. This procedure should be
followed to the extent practical.
C6. Accredited Technical Competence
Anyone performing work within the EPA traceability scheme must provide proof of technical
competence by meeting minimum training requirements. Training records and plans must be
documented showing initial and ongoing training. Inter- and intra- laboratory comparisons and
outside accreditation should also be considered.
C7. Measurement Assurance
EPA incorporates measurement control steps to ensure the validity of the O3 verification process.
The procedures and measurands are defined throughout this TAD. EPA recommends that all
laboratories verifying O3 transfer standards:
1. Maintain a transfer standard designated as a check standard for generating control charts
to ensure the standard of higher authority is in control;
2. Conduct additional redundant checks and verifications;
3. Conduct comparisons by sending a transfer standard to other branches in the traceability
chain to ensure consistency among operators;
4. Conduct frequent reviews of existing QC data collected at monitoring sites;
5. Develop a corrective action plan for when transfer standards are not within established
parameters.
24 https://www.nist.gov/svstem/files/documents/2019/05/13/sop-l-calibration-certificate-preparation-20190506.pdf
C-3
-------
Appendix D Rationale and Testing Methodology for O3
Verification and Re verification Acceptance Criteria
Dl.Introduction
In revising the 2013 version of Transfer Standards for the Calibration of Air Monitoring
Analyzers for O3, EPA examined every step in the transfer standard verification and
reverification process and examined the acceptance criteria used in each process. EPA was
interested in determining if these processes or acceptance criteria needed to be updated. This
appendix provides a general overview and a plain language description of EPA's approach when
reviewing and changing verification/reverification processes and acceptance criteria.
D2. Overview of Verification and Reverification Procedure
Upon receipt of a new candidate transfer standard, and then following any adjustment or major
repair of the device, the device is subjected to formal verification testing. In verification testing,
the candidate transfer standard's reported measurements are compared to those generated by a
standard of higher authority at the same concentration test point. A verification test consists of a
minimum of three stable test cycles, each consisting of at least six concentration points
distributed across the range of O3 concentrations expected to be encountered during normal use,
as well as 0 ppb. These concentration points should be as evenly spaced as possible across the
range and could also include the NAAQS value of 70 ppb. Once a transfer standard has been
successfully verified, reverifications are required at a set frequency depending upon their use.
D3.Could the minimum number of test cycles within a verification test be reduced from six
to three?
EPA performed a statistical analysis to determine whether a statistically significant difference in
the outcome would occur if the number of cycles in a verification test was reduced from six to
three. When performing this analysis on Level 3 verification test data when data for only either
the first 3 or last 3 cycles were considered, no significant differences in the intercepts or slopes
were observed among cycles at a 0.05 level. Thus, EPA determined that the minimum number of
test cycles could be reduced from six to three.
D4.Acceptance criteria for individual measures at specific test concentration points.
Within each concentration test point within a test cycle, the measurements taken by the candidate
transfer standard are paired with the measurements taken by the standard of higher authority. The
two measurements within a sampling pair are first assessed to ensure they do not differ
significantly from each other. The measurements within a pair must not differ by more than the
following thresholds:
• 3.1%, for concentration points above 50 ppb. (Here, the difference between the two
paired measurements is divided by the measurement for the standard of higher authority
and expressed as a percentage. The sign of the difference is ignored.)
D-l
-------
• 1.5 ppb in absolute value, for concentration points at or below 50 ppb. (Here, only the
difference between the two paired measurements is considered, and the sign of the
difference is ignored.)
These acceptance criteria were determined in earlier versions of the TAD.
If any concentration test point fails its appropriate acceptance criterion, data from that test cycle
are discarded and the test cycle is repeated.
D5.Acceptance criteria for cycle-specific values of the slope and intercept estimates within
a verification test.
Once the acceptance criteria for individual measurement pairs have been achieved within a given
test cycle, a straight line is fitted to the paired measurements across all tested concentration
points within the cycle (using "least squares regression") to predict the measurement from the
candidate transfer standard given the measurement from the standard of higher authority. (Thus,
the amount of data used to determine a cycle's straight-line relationship corresponds to the
number of concentration points, including 0 ppb, in the given test cycle.) The fitted straight line
is defined by its estimated intercept and slope:
• The intercept (in ppb) corresponds to the measurement value for the candidate transfer
standard which the line predicts when
the measurement for the standard of
higher authority equals 0 ppb.
o For each of the three test
cycles, the intercept estimate
must fall between -3 ppb and
+3 ppb.
• The slope (which is unitless)
corresponds to the predicted change in
the candidate transfer standard's measurement value that occurs for each 1.0 ppb increase
in the standard of higher authority's measurement value.
o For each of the three test cycles, the slope estimate must fall between 0.97 and
1.03.
These acceptance criteria were determined in earlier versions of the TAD.
If either the slope estimate or the intercept estimate within a given test cycle falls outside of its
respective acceptance criterion, then the results for the candidate transfer standard do not
sufficiently agree (or coincide) with the results for the standard of higher authority for that cycle,
and therefore the results are discarded and the cycle must be repeated.
D6.Could the three test cycles within a verification test be performed on the same day, or
must they be performed on different days?
EPA performed a statistical analysis to assess whether the three test cycles within a verification
test should continue to be performed on different (consecutive) days, or if they could be
Authoritative Standard Concentration
D-2
-------
performed on the same day. Performing test cycles on the same day could lead to resource
savings as the time needed to invest in performing a full verification test would be reduced, and
the daily set-up for testing would be reduced from three days to one day. In this statistical
analysis, EPA wished to assess, for a typical transfer standard, how variability (in the slopes and
intercepts of the fitted lines) among cycles performed on a single day compared to the variability
across cycles performed on different days. The results of the statistical analysis suggested that
the variability in the intercept and slope estimates, calculated across cycles, tended not to differ
significantly if those cycles were performed within the same day or performed on different days.
(In this analysis, six cycles were assumed within a verification test, rather than three.) Having
each cycle performed on the same day was not expected to result in a much higher or lower
likelihood of exceeding acceptance criteria compared to performing one cycle per day. Thus,
within a verification test, EPA decided to permit the three test cycles to be performed within the
same day, as long as conditions were in place to minimize the extent to which the outcomes from
one cycle would influence the outcomes from another cycle.
D7.Acceptance criteria for the standard deviation of the slope and intercept estimates
across cycles within a verification test.
Once the acceptance criteria for the three intercept estimates and the three slope estimates are
achieved within a verification test, the stability of these estimates across the three cycles was
confirmed. A "standard deviation" was used as the measure of stability in these estimates. The
standard deviation is a measure of how spread out the estimates are, and in particular, the extent
to which they tend to deviate from their mean value.
The direction given earlier in this document is that all three cycles of a verification test can be
performed within a single day. Therefore, the acceptance criteria for the standard deviation of the
three slope estimates and the standard deviation of the three intercept estimates were derived
using historic test data for those verification tests in which the three test cycles occurred within
the same day, and the first of the three cycles was the first cycle tested on that day. These data
were made available from the EPA Regions for select Levels 2 and 3 transfer standards. Note
that it is assumed that this set of available data well represent the full set of all possible outcomes
from the single-day verification of Levels 2 and 3 transfer standards.
The acceptance criteria on the standard deviations of the slopes and intercepts took the form of a
threshold value (one for the slope estimates, and one for the intercept estimates) against which
the standard deviations were compared. If a standard deviation exceeded its threshold, then the
standard deviation was deemed too large, and thus the estimates were not adequately stable
across the three test cycles. The statistical analysis which EPA used to derive the acceptance
thresholds needed to accurately portray the distribution of standard deviations in the slope and
intercept estimates (especially the part of the distribution representing large values). Because the
available data to inform this distribution was quite limited, EPA applied a statistical simulation
procedure to the available data when generating the distribution and deriving the acceptance
threshold from it. This simulation required two confidence levels to be selected:
D-3
-------
• A confidence level on the upper bound on the standard deviation within a verification
test. (That is, the level of confidence at which we are willing to assume that the likelihood
of the true standard deviation being below a specified value for a typical verification
test.) EPA picked this confidence level to be 90%.
• A confidence level on the upper bound on the standard deviation across multiple
verification tests. (That is, the level of confidence at which we are willing to assume that
the likelihood of the true standard deviation of the test-specific mean estimate across
cycles being below a specified value across the verification tests.) EPA picked this
confidence level to be 95%.
Based on these settings, the statistical analyses yielded acceptance thresholds of 0.0075 for the
standard deviation of slope estimates, and 1.00 for the standard deviation of intercept estimates.
If for a given verification test, the standard deviation of either the slope estimates or the intercept
estimates exceed its respective threshold, then the outcomes are not considered stable among the
test's three cycles, and the verification test must be repeated on the given candidate transfer
standard.
D8.Acceptance criteria for reverification testing).
Initially, EPA considered an approach for determining acceptance criteria for reverification
testing which involved obtaining data from the standard's last three test cycles that passed all
acceptance criteria and using the data to calculate "95% prediction intervals" on the slope and
the intercept of the anticipated regression line for a future (reverification) testing cycle applied to
that standard (hence use of the term "prediction"). The estimated slope and intercept of the
reverification's test cycle needed to fall within their respective prediction intervals (for the same
standard) to achieve the acceptance criteria for reverification. In testing this approach on
example verification and reverification data, however, the prediction intervals were very narrow
for standards having high repeatability. Therefore, this approach led to concerns that it would
declare highly reliable standards as failing their reverification tests with high likelihood. This
failure could also be caused by the small nominal drift in measurements that normally occurs
over the elapsed time.
To avoid the ramifications caused by these concerns, EPA rejected this initial approach and took
a different approach for determining acceptance criteria for reverifications. The adopted
approach utilizes data from the candidate transfer standard's most recent successful verification
test - note that this may be different from the sampler's three most recent successful test cycles,
as some may have been reverification cycles done on different days. Recall that verification
testing involves performing three independent test cycles, each performed on the same day. As
given by Equations 8 and 9 of Appendix A, the standard deviation of the three slope estimates
(SDm) and the three intercept estimates (SDb), respectively, are calculated from the three fitted
regression lines generated within a verification test. For a series of historic verification tests
performed across a set of multiple transfer standards, EPA characterized the distributions of the
standard deviations (for both SDm and SDb) calculated within these tests. Of particular interest
were the medians and 95th percentiles of these two distributions. From these relationships and
D-4
-------
from other data investigations, the following two acceptance criteria were established for the
estimated slope and intercept of the reverification test cycle data:
• The slope of the regression line fitted to the reverification cycle data must be within
±0.015 of the value of the mean slope (fn in Equation 6 in Appendix A) associated with
the most recent successful verification test.
• The intercept of the regression line fitted to the reverification cycle data must be within
±1.5 ppb of the value of the mean intercept (b in Equation 7 in Appendix A) associated
with the most recent successful verification test.
Note that these acceptance criteria for reverifications are one-half (50%) of the acceptance
criteria for the mean slope and mean intercept from the verification test, as specified in Table 4-
1, Section 4.4.1, and Section D.5. These acceptance criteria take into account the naturally
occurring drift in measurements that occurs over time.
D-5
-------
Appendix E Qualification Process
(reprintedfrom "Transfer Standards for the Calibration of Ambient Air Monitoring Analyzers
for Ozone, 2013 Information on 03 generators deleted)
The first step in establishing the authority of a candidate transfer standard is to prove that it
qualifies for use as a transfer standard. In other words, can the output (either an actual O3
concentration or a concentration assay, depending on the type) of the candidate transfer standard
be trusted under the changing conditions of use that might be encountered in field use.
Qualifying new instruments can help ensure the instruments will perform under various field
conditions that might not get observed during a traditional verification processes described in
Section 4 of this document.
The primary requirement of a transfer standard is repeatability - repeatability under the stress of
variable conditions that may change between verification and use. A candidate transfer standard
is qualified by proving that it is repeatable over an appropriate range for each variable likely to
change between the time and place of verification and the time and place of use. According to
the specifications in Section 3, the repeatability must be within ± 4% or ± 4 ppb, whichever is
greater, for each condition or variable that may change between the point of verification and the
point of use.
Selecting the conditions that are likely to vary and that may affect the repeatability of the device
or procedure is largely a matter of intelligent, informed, judgment. To a large extent, the
variables will depend on the nature of the device or procedure; for some candidate transfer
standards, the variables to be considered may be quite numerous. It is the user's responsibility to
determine all of the conditions to be considered in the demonstration of repeatability before a
candidate transfer standard can be considered qualified for use as a transfer standard. Common
conditions likely to affect a wide variety of types of transfer standards include such items as
ambient temperature, line voltage, barometric pressure, elapsed time, physical shock, etc. These
variables are discussed individually later in this section. Conditions not likely to affect the
transfer standard can usually be eliminated from consideration. The user must, however, be
constantly alert for the unusual situation where an unexpected condition may significantly affect
the repeatability of a transfer standard.
Note that a transfer standard does not necessarily need to be constant with respect to these
variables, only repeatable or predictable. While it is certainly desirable that a device or
procedure be insensitive to any given variable, it may still qualify as a transfer standard if it is
repeatable. For example, it may be difficult to find or design a generation-type transfer standard
device that is insensitive to barometric pressure. However, if it is repeatable with respect to
barometric pressure, the relationship can be quantitatively defined by a curve or table. At the
time of use, the local barometric pressure must be measured, and the curve or table used to
"correct" the transfer standard's indicated output. This technique is acceptable for one or
perhaps two variables. But beyond two variables, the difficulties of determining and specifying
E-l
-------
the relationship to the variables may become impractical. Fortunately, sensitivity to most
variables can be reasonably controlled.
Demonstration of repeatability for a candidate transfer standard normally requires testing for
each condition that could or may affect it. Typical tests for common conditions are discussed
below. Again, intelligent judgment is required to determine what conditions to test and the
extent of testing required to qualify the device or procedure. For commercially available transfer
standard devices, some or all the testing may be carried out by the manufacturer, thereby
reducing the burden on the user. In some cases, it may be possible to judiciously substitute
design rationale for actual testing. For example, a device whose power supply is designed to be
highly regulated electronically may not require specific line voltage tests. However, such
situations should be viewed with considerable skepticism.
The preceding discussion brings up the further question of whether candidate transfer standards
must be tested individually or whether they can be qualified by type, model, or agency. The
units of commercially produced transfer standard devices are designed and manufactured to be
identical and should therefore have very similar characteristics. The manufacturer could carry
out the necessary qualification tests on representative samples, sparing the user the burden of
testing each unit or the cost of paying the manufacturer to test each unit individually. Under this
concept, it would certainly be appropriate to require the manufacturer to guarantee that each unit
meets appropriate performance specifications. However, the user should assume a skeptical
attitude, in view of manufacturing tolerances and possible defective components, and carry out at
least some minimal tests to verify that each unit is acceptable.
In the case of unique devices assembled by users, testing for all pertinent conditions which could
or might affect the device are normally required.
QUALIFICATION TESTS
Some of the more common conditions likely to be encountered or to change while using transfer
standards and that may often affect the repeatability of the device or procedure are discussed
below. Also discussed are ways or approaches to test for sensitivity to the condition. As noted
previously, the exact conditions or variables that must be considered depend on the specific
nature of the device or procedure. The user (or manufacturer, etc.) should determine the
conditions for each case on an intelligent judgmental basis derived from a complete
understanding of the operation of the device or procedure and supported by appropriate rationale.
Once the conditions to be considered have been determined, the objective of the qualification
tests is either a or b:
a) to demonstrate that the candidate transfer standard's output is not affected by more than ± 4%
or ± 4 ppb (whichever is greater) by the condition over a range likely to be encountered during
use of the device or procedure;
b) to demonstrate that the candidate transfer standard's output is repeatable within ± 4% or ± 4
ppb (whichever is greater) as the variable is changed over a range likely to be encountered during
use, and to quantify the relationship between the output and the variable.
E-2
-------
Temperature
Changes in ambient temperature are likely to occur from place to place and from one time to
another. Temperature changes are very likely to affect almost all types of transfer standards
unless appropriate means are used to avoid adverse effects. Temperature affects transfer
standards in many ways: changes in the action of components, changes in chemical reactions or
rates of reaction, volume changes of gases, electronic drift, variable warm-up time, etc. The
most important effects may well be (1) changes in the output of generation devices, (2) changes
in the sensitivity of O3 analyzer-type systems, and (3) changes in the volume of air flows which
must be measured accurately.
Temperature effects can be minimized in several ways. Since shelters for ambient air monitoring
are normally maintained at about 20 - 30° C (with some flexibility for fluctuations) all transfer
standards should be proven to be repeatable within this range. Transfer standard devices may be
made insensitive to temperature changes by design, such as thermostatic regulation of sensitive
components or of the entire device, or by temperature compensation.
Temperature effects on air flow measurement can be minimized by the use of mass flowmeters,
which do not measure volume, or by the regulation of gas temperatures. In another approach,
ordinary ideal-gas-law corrections could be made manually to adjust to measured volumetric
flowrates. However, when using orifice control or measurement devices such as critical orifices
and rotameters, be sure to use an appropriate correction formula.
Testing a candidate transfer standard for sensitivity to temperature is facilitated by the use of a
controlled temperature chamber. However, successful temperature tests can be carried out in
many ordinary laboratories where the temperature can be manually controlled by adjusting
thermostats, blocking air vents or outlets, opening doors or windows, or using supplemental
heaters or air conditioners. A reasonable temperature range would be 20 to 30°C (68 to 86°F).
Broader temperature ranges could be used if appropriate.
The candidate transfer standard is tested by comparing its output to a stable concentration
reference. This reference should ideally be a UV analyzer system at least one level above the
level of the transfer standard. It would be best to locate the reference outside of the variable
temperature test area. The candidate transfer standard should be tested at a minimum of 3
different points over the temperature range, including the extremes, and at a minimum of 3
E-3
-------
0.5
0.45
0.4 -
0.35 -
I °-3
g* 0.25
i 0-2
0.15
0.1
0.05
±0.018 4%
+ 0.01 4%
± 0.004 4ppb
15
20
25
30
Temper.itwe. Deg C
35
40
45
Figure B.l Example of temperature qualification test results shov ing no
dependence on temperature.
different concentrations. Be sure
to allow sufficient time for the
device or any instruments or
equipment associated with the
transfer standard to equilibrate
(min time period or until reading
stabilized) each time the
temperature is changed. The test
results should be plotted in a
fashion similar to the example
shown in Figure B.l.
If the candidate transfer standard
has a significant temperature
dependence, additional test points
at various concentrations and
temperatures should be taken to
define the relationship between
output and temperature accurately.
Furthermore, if the candidate turns
out to have a dependence on more
than one condition or variable,
tests must be carried out over the
range of both variables
simultaneously to determine any
interdependence between the two
variables. Once the test data are
acquired, they should be analyzed
to determine if some general
formula or curve can be derived
(either analytically or empirically)
to predict the correct O3
concentration at any temperature
in the range (see Figure B.2). The correction formula or curve must be accurate within ± 4% or
± 4 ppb, whichever is greater. If two or more variables are involved, a family of curves may be
required; unless the relationship is rather simple, this situation may prove impractical in actual
use.
30 35
Temperature, Deg C
Figure B.2 Example of a temperature dependence quantitatively
defined correction factor.
Line Voltage
Line voltage is very likely to vary from place to place and from one time to another. Good
electrical or electronic design of the transfer standard should avoid sensitivity to line voltage
variations, but poorly designed equipment can easily be affected. In addition, line voltage
sensitivity may appear only as long-time thermal drift, a rather subtle effect.
E-4
-------
Aside from adequate design, line voltage effects can be minimized by the addition of an outboard
line voltage regulator. However, such devices may distort the line voltage waveform, thereby
adversely affecting some types of equipment. If such regulators are used, it is important that the
same regulator is used during both verification and use of the transfer standard. Restriction of
the transfer standard to a line voltage range in which the effects are insignificant is another
alternative, but that would require monitoring the voltage during use and may preclude use at
some sites.
Testing for line voltage sensitivity can be carried out along the same lines as described for
temperature testing. The line voltage can be varied by means of a variable voltage transformer
("Variac") and measured by an accurate ac voltmeter. Do not use electronic "dimmer" controls
which operate on a delayed-conduction principle, as such devices cause drastic waveform
distortion.
A line voltage range of 105 to 125 volts should adequately cover the vast majority of line
voltages available in the U.S. If the transfer standard is used when powered by a small power
generator, it should be checked for frequency dependence.
Barometric Pressure/Altitude
Since O3 concentrations are gaseous in nature, all transfer standards will probably have some
basic or inherent sensitivity to change in barometric pressure. Unfortunately, it is rather difficult
to minimize barometric pressure effects by design. Air pressures can be regulated mechanically
against an absolute reference, but most such schemes are not practical when working with O3
concentrations because of restrictions to inert materials such as glass or Teflon. At a constant
altitude, normal day-to-day variation in barometric pressure is only a few percent. If the use of
the transfer standard can be restricted to altitudes within a hundred meters of the verification
altitude, it may be acceptable to neglect the barometric effect entirely. However, if the use of a
transfer standard is necessary at altitudes significantly different than the calibration altitude, then
pressure effects cannot generally be ignored.
E-5
-------
Although not readily preventable, pressure effects are likely to be repeatable. As a result,
barometric pressure may be the variable most likely to be handled by the defined-relationship
approach discussed previously in
connection with temperature
effects. The technique is very
similar to the technique used to
determine a temperature
relationship; hopefully, a unique
quantitative relationship will
result, such as that illustrated in
Figure B.3. Remember that in
any work with O3 concentrations
at altitudes significantly above
sea level, the
concentration units must be
clearly understood. The volume
ratio concentration units must be clearly understood. The volume ratio concentration units (ppm,
ppb, etc.) are independent of pressure, while density units such as |ig/m3 are related to pressure.
(However, the |ig/m3 unit defined and used by EPA is "corrected" to 1.01 kPa (760 mm Hg) and
25°C and is therefore related to ppm by a constant.)
Testing with respect to barometric pressure may be difficult. The use of a variable pressure
chamber is the best approach, but few laboratories have access to such facilities. It is
conceivable that various pressures could be obtained in a manifold setup, but construction of
such an apparatus is difficult and of questionable validity. The use of a mobile laboratory
vehicle which can be driven to various altitudes to conduct tests may offer the most feasible
solution. Some types of transfer standards may not require pressure tests because their pressure
sensitivity is well known. Some analyzer-type devices are clearly related directly to gas
density, where a simple ideal-gas-law correction can be applied. Pressure tests are not needed
for these types. For commercially-produced devices, the manufacturer would be expected to
carry out the necessary qualification tests and to offer the devices as type-approved, at least with
respect to pressure effects.
As a final note of encouragement, automatic compensation for barometric pressure is rapidly
becoming economically feasible for some types of O3 transfer standards by the incorporation of
microprocessor technology. At least two manufacturers have used this approach in
commercially available instruments.
Elapsed Time
As the elapsed time between verification and use increases, the confidence in the repeatability
decreases. As a result, periodic reverification is needed. Some types of O3 generation devices
have a definite loss of output (decay) with time. This decay is usually associated with use-time
or on-time rather than total elapsed time. Since the decay rate tends to be quantifiable, it can be
1.3 ¦
S 1
0.4-
Pauabolic Curve
Y"—*¦ A+BX+CXXX
A * 3.068399
B—~ ¦ 005191
000003
COKJR. COEF —»¦ 999912
616 «sa no ?eo
H.romctfie pre»uir, MM
Figure B.3 Example or st defined barumelric pressure dependence
E-6
-------
accommodated with the defined-relationship mechanism discussed in connection with
temperature effects: the transfer standard is equipped with an hours meter or another time
measuring device and a series of tests over a sufficient time period can then be used to determine
the decay rate. During use, a correction to the output is applied based on the number of hours of
on-time since the last verification.
Another approach is to recertify such a transfer standard often enough so that the error due to
decay never exceeds the ± 4% or ± 4 ppb specification.
Variability
The adequacy of the relationship between a transfer standard and a authoritative reference O3
standard is dependent on the variability of the transfer standard. Variability reduces confidence
in transfer standards. A high degree of variability may be cause for disqualifying a device or
procedure for use as a transfer standard, or for selecting one with lower variability. Although the
verification procedure in Section 4 includes a test for variability, more extensive tests for
variability may be necessary to qualify a transfer standard because the verification test is for
variability in the slope of the verification relationship and not for individual point variability.
Furthermore, variability may be due to changes in conditions not encountered during
verification.
Many different types of transfer standards may have excessive variability for a variety of
reasons. Qualification variability testing is perhaps most needed to test for the effect of a variety
of non-specific or non-quantitative variables that cannot be tested individually. Whenever
increased variability can be assigned to a specific cause, corrective actions or restrictions can be
and should be applied to reduce the variability.
The variability test should be carried out on a single-point basis. A series of at least 6 single-
point comparisons should be made between the candidate transfer standard and a UV reference at
each of at least two fixed concentrations - one low concentration (less than 0.1 ppm) and one
high concentration (over 80% of the upper range limit). These comparisons should be made over
a variety of conditions and situations and over a number of days. For each concentration, verify
that all O3 concentration measurements determined by the UV reference standard are very nearly
equal. Then calculate the average of the 6 (or more) concentrations indicated by the transfer
standard, using the following equation:
Ave = -Jy1
n i=i
Where: n = number of comparisons
yi = O3 concentration indicated by the transfer standard
Determine the difference between each concentration indicated by the transfer standard and the
average concentration (yi - Ave). Each difference must be less than ± 5% of the average (for
concentrations over 0.1 ppm) or less than ± 5 ppb (for concentrations less than 0.1 ppm).
E-7
-------
For this test, the acceptable limits are ± 5% or ± 5 ppb rather than ± 4% or ± 4 ppb, because the
test is for general variability, which may derive from a number of non-identifiable causes. Under
these circumstances slightly wider limits than those allowed for the other qualification tests are
acceptable.
One technique that can reduce variability and improve accuracy is repetition and averaging. For
example, the variability of assay procedures can be reduced by assaying each concentration
several times and averaging the results. Of course, if this technique is used, it becomes a
necessary part of the transfer standard procedure and must be carried out each time the transfer
standard is used and certified.
Relocation
A transfer standard obviously needs to maintain repeatability after being moved and possibly
encountering mechanical shocks, jolts, and stress. Any electrical or thermal stress incident to
turning the device or equipment on and off frequently is also of concern, as is consideration of
orientation or set-up factors.
Tests for these conditions, while perhaps not particularly quantitative, should include actually
moving the candidate device or equipment to different locations and comparing the output each
time it is returned. Tests could also include mild shock or drop tests, or tests for any set-up
factors which can be specifically identified, e.g., physical orientation, removal of covers, any set-
up variations. Any cause-and-effect-relationship discovered should be investigated completely.
The tests may be conveniently combined or included with those discussed previously for
variability.
Operator Adjustments
Those transfer standard devices whose output is to be related to an operator adjustment should be
tested for repeatability with respect to the adjustment. Mechanical adjustments might need to be
tested for play, backlash, hysteresis, slippage, and resolution. Other types of adjustments may
require tests for analogous aspects. If possible, specific tests should be used. For example,
approaching a given setting from both above and below the setting might be appropriate for
testing play or hysteresis. If specific tests cannot be designed, then simple repeatability tests at
several different settings should be carried out.
Malfunctions
The usefulness of a transfer standard is dependent on the degree of confidence that can be put on
its ability to reproduce O3 standards. While any device is subject to occasional malfunctions,
frequent malfunctions would certainly compromise the purpose of a transfer standard. Of
particular concern are non-obvious type malfunctions that can cause a significant error of which
the operator is unaware. While no specific tests for malfunctions are normally used, the tests
described above for the other conditions need to be repeated periodically to check for non-
obvious malfunctions. After a malfunction has been corrected, the transfer standard must be
recertified.
E-8
-------
Other Conditions
Any other condition that might affect a candidate device or procedure or that might cause change
between the point of verification and the point of use should be tested.
E-9
------- |