vvEPA
       Quality Assurance Guidance
       Document 2.12
        Monitoring PIVb.s in Ambient Air
        Using Designated Reference or
        Class I Equivalent Methods

-------
Page Intentionally Left Blank

-------
                                             EPA-454/B-16-001

                                                 January 2016
        Quality Assurance Guidance Document 2.12
Monitoring PM2.5 in Ambient Air Using Designated Reference
              or Class I Equivalent Methods
         U.S. Environmental Protection Agency
      Office of Air Quality Planning and Standards
            Air Quality Assessment Division
                   RTP,NC27711
                    January 2016

-------
                                       Disclaimer

The statements in this document, with the exception of referenced requirements, are intended solely
as guidance. This document is not intended, nor can it be relied upon, to create any rights enforceable
by any party in litigation with the United States. EPA may decide to follow the guidance provided in
this document, or to act at variance with the guidance based on its analysis of the specific facts
presented. This guidance may be revised without public notice to reflect changes in EPA's approach
to implementing 40 CFR Part 50, Appendix L and 40 CFR Part 58, Appendices A and B.

Mention of commercial products or trade names should not be interpreted as endorsement. Some
types of instruments currently in use may be described in text or in example figures or tables.
Sometimes these products are given as a typical and perhaps well-known example of the general class
of instruments. Other instruments in the class are available and may be fully acceptable.

-------
2.12. Table of Contents                                              January 2016    Page ii
 Table of Contents
 Section                                                                          Page

            Disclaimer	 i
            List of Figures	 vii
            List of Tables	viii
            List of Acronyms and Abbreviations	ix

   1.0      Scope and Applicability	1 of 6
            1.1   Overview and Purpose	1
            1.2   Applicability	1
            1.3   Conventions	2
            1.4   Format and Structure	2
            1.5   Overview of Reference and Equivalent Methods for PM2.5	3
                  1.5.1  Reference Methods	3
                  1.5.2  Equivalent Methods	4
            1.6   Limitations of PM2.5 Reference and Class I Equivalent Methods	5

   2.0      Prerequisites	1 of 8
            2.1   Overview and Definitions	1
            2.2   Personnel Qualifications	1
                  2.2.1  Laboratory Personnel Qualifications	1
                  2.2.2  Field Personnel Qualifications	6
                  2.2.3  Training for Laboratory and Field Personnel	6
            2.3   Health and Safety Warnings	7
            2.4   Cautions	8

   3.0      Summary	1 of 5
            3.1   Overview	1
            3.2   Illustrations of Sampler Inlet Components and Sample
                  Flow Paths	2
            3.3   Summary of Field and Laboratory QC Checks	3
            3.4   Summary of Sampling Procedures for PM2.5	3

   4.0      Procurement of Equipment and Supplies	1 of 10
            4.1   Overview	1
            4.2   Procurement Prerequisites—Field Operations	2
                  4.2.1  Reference or Equivalent Method Sampler	2
                  4.2.2  Calibration Equipment	4
                  4.2.3  Flow Verification CheckDevice	4
                  4.2.4  Audit Equipment	5
            4.3   Procurement Prerequisites—Laboratory Operations	5
                  4.3.1  Filter Media	5
                  4.3.2  Filter Support Cassettes	7
                  4.3.3  Filter Cassette Protective Containers	7

                                       Method 2.12-PM25 Reference/Class I Equivalent Methods

-------
Table of Contents (continued)
2.12. Table of Contents                                              January 2016     Page iii

                  4.3.4  Insulated Shipping Container	8
                  4.3.5  Filter Handling Containers	8
                  4.3.6  Analytical Microbalance	8
                  4.3.7  Calibration Reference Standards	9
                  4.3.8  Temperature and Relative Humidity Logger	10
                  4.3.9  Audit Equipment	10

   5.0      Field Installation	1 of 4
            5.1   Siting Requirements	1
                  5.1.1  Spatial and Temporal Scales of Site	1
                  5.1.2  Sampler Location	1
                  5.1.3  Safety	2
                  5.1.4  Electrical Considerations	2
                  5.1.5  Security	2
            5.2   Sampler Installation Procedures	2
                  5.2.1  Receipt of Sampler	2
                  5.2.2  Initial Evaluation of Samplers	3
                  5.2.3  Setup at Sampling  Site	3
                  5.2.4  Field Evaluation	4

   6.0      Calibration Procedures	1 of 18
            6.1   Overview	1
            6.2   General Aspects of Flow Rate Measurement and PIVh.s
                  Sampler Calibration	3
            6.3   Calibration of the Sampler Flow Rate Measurement
                  System	4
                  6.3.1  General Requirements and Guidance	5
                  6.3.2  Flow Rate Calibration Standards	7
                  6.3.3  NIST Traceability  and Certification of Flow Rate Standards	8
                  6.3.4  Outline of Generic Flow Rate Calibration Procedure	9
            6.4   Calibration of Sampler Temperature Sensors	10
                  6.4.1  General Requirements and Guidance	10
                  6.4.2  Temperature Calibration Standards	11
                  6.4.3  NIST Traceability  and Certification of Temperature Standards	12
                  6.4.4  Outline of Generic Temperature Calibration Procedure	12
            6.5   Calibration of Sampler Barometric Pressure Sensor	14
                  6.5.1  General Requirements	14
                  6.5.2  Outline of Generic BP Calibration Procedures	14
                        6.5.2.1   Digital Pressure Indicator	15
                        6.5.2.2   Fortin-Type Barometer Readings	16
                        6.5.2.3   Aneroid-Type Barometer	16
            6.6   Leak Checks	16
            6.7   Verification/Calibration Frequency	16
                                       Method 2.12-PM25 Reference/Class I Equivalent Methods

-------
Table of Contents (continued)
2.12. Table of Contents                                             January 2016    Page iv
   7.0      Field Operation	1 of 17
            7.1  Overview	1
            7.2  Activities to Perform Each Site Visit	1
                 7.2.1   Beginning a Run	2
                 7.2.2   Ending a Run	5
                 7.2.3   Sample Validation	8
                 7.2.4.  Sample Handling	9
                        7.2.4.1   Handling a Valid Sample	9
                        7.2.4.2   Handling a Questionable Sample	10
            7.3  Activities to Perform Every Five Operating Days	10
                 7.3.1   Impactor Well Cleaning	10
            7.4  Activities to Perform Every Four Weeks	10
                 7.4.1   VSCC™ Cleaning	11
                 7.4.2   Field Quality Control Equipment and Standards	11
                 7.4.3   External Check	11
                 7.4.4   Internal Filter Bypass Leak Check	14
                 7.4.5   Procedure for Temperature Verification Check	14
                 7.4.6   Procedure for Barometric Pressure Verification Check	15
                 7.4.7   Procedure for Field Flow Rate Verification Check	16

   8.0      Sampler Maintenance	1 of 6
            8.1  Overview	1
            8.2  Five-Day Maintenance Procedures	2
                 8.2.1   Water Collector	2
                 8.2.2   WINS Impactor Well	2
            8.3  Monthly Maintenance Procedures	3
                 8.3.1   Inlet	3
                 8.3.2   WINS Impactor Housing and Well	4
                 8.3.3   Cleaning of VSCC™ Type PM2.5 Separators  	4
                 8.3.4   Other	5
            8.4  Quarterly Maintenance	5
                 8.4.1   Downtube	5
                 8.4.2   Impactor	5
                 8.4.3   Other	6
            8.5  Other Periodic Maintenance	6
            8.6  Refurbishing PM2.5 Samplers	6

   9.0      Gravimetric Laboratory Design and Setup	1 of 14
            9.1  Overview	1
            9.2  Weighing Room	1
            9.3  Microbalance	3
            9.4  Logging System	4


                                       Method 2.12-PM25 Reference/Class I Equivalent Methods

-------
Table of Contents (continued)
2.12. Table of Contents                                             January 2016     Page v

            9.5  Data Management System	6
            9.6  Static Neutralization	7
            9.7  Mass Reference Standards	9
                 9.7.1  Certification Procedures-Primary Standard	9
                 9.7.2  Certification Procedures-Working Standard	11

10.0        Filter Preparation and Analysis	1 of 23
            10.1 Overview	1
            10.2 Filter Handling	1
            10.3 Filter Integrity Check	2
            10.4 Filter Conditioning	3
            10.5 Quality Control Blanks	9
            10.6 Pre-Sampling Filter Weighing (Tare Weight)	14
            10.7 Post-Sampling Documentation and Inspection	18
            10.8 Post-Sampling Filter Weighing (Gross Weight)	20
            10.9 Calculation of Net Mass Filter Loading	21
            10.10 Internal Quality Control	22
            10.11 Filter Archive Requirements	22

 11.0       Performance Evaluation Procedures	1 of 15
            11.1 Overview	1
            11.2 Performance Evaluations	2
                 11.2.1  Sampler Flow Rate Audit	3
                        11.2.1.1  Flow Rate Audit Data Calculations	5
                 11.2.2  Temperature Audit Procedure	5
                 11.2.3  Barometric Pressure Audit Procedures	7
                 11.2.4  Time Audit	7
                 11.2.5 Assessment of Precision Using Collocated Measurements	7
                        11.2.5.1  Procedure for Collocated Measurements	7
                 11.2.6  Federal Reference Method Performance Evaluation	8
                 11.2.7  Balance Accuracy Assessment	9
                 11.2.8  Temperature and Relative Humidity Logger Audit Procedure	9
            11.3 Systems Audits	10

    12.0     Calculations, Validations, and Reporting of PMi.s Monitoring
            Data	1 of 7
            12.1 Overview	1
            12.2 Calculations	1
                 12.2.1  Sample Volume Calculations	1
                 12.2.2  Net PM2.5 Mass Calculation	2
                 12.2.3  PM2.5 Concentration Calculation	2
            12.3 Verification of Manual Calculations and Data Entry	3
                 12.3.1  Verification of Manual Calculations	3
                                       Method 2.12-PM25 Reference/Class I Equivalent Methods

-------
Table of Contents (continued)
2.12. Table of Contents                                            January 2016    Page vi

                 12.3.2 Verification of Manual Data Entry	3
            12.4 Validation	4
                 12.4.1  Data Validation	4
                 12.4.2  Validation of Software Used to Process PM2.5 Data	6
            12.5 Data Reporting	7
                 12.5.1 Rounding	7
                 12.5.2 Rounding Rules for Comparison with NAAQS	7

   13.0     Data and Records Management	1 of 6
            13.1 Overview	1
            13.2 Methodology for Data and Records Management	1
            13.3 PM2.5 Records to Create and Retain	2
                 13.3.1 Sampler Siting and Maintenance Records	2
                 13.3.2 Analytical Laboratory Installation Records	3
                 13.3.3 Field Sampling Operation Records	3
                 13.3.4 Weighing Laboratory Operation Records	3
                 13.3.5 Quality Assurance Records	4
            13.4 Data Reporting Requirements	4

   14.0     Assessment of Measurement Uncertainty for Monitoring Data	1 of 6
            14.1 Overview	1
            14.2 Flow Rate Audits	1
            14.3 Bias Assessment	2
            14.4 Precision	3
                 14.4.1 Overview	3
                 14.4.2 Number of Collocated Samplers Required	4
                 14.4.3 Location of Collocated Samplers	5
                 14.4.4 Schedule of Operation of Collocated Samplers	6

   15.0     References	1 of 3
 Appendix A—Measuring Alkalinity of Filters	1 of 12
                                      Method 2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. List of Figures and Tables                                        January 2016    Page vii


 List of Figures

 Number                                                                                Page

    3.1       PM2.5 inlet head	2 of 5
    3.2       Air flow through the PM25 impactor well and filter holder	2 of 5
    3.3       Example of a very sharp cut cyclone	2 of 5

    4.1       Example procurement log	2 of 10
    4.2       Example Equipment for transporting filter cassettes	7 of 10

    6.1       Example Calibration Work Sheet	18 of 18

    7.1       Example of a PM2 5 field datasheet	4 of 17
    7.2       Example of a filter chain-of-custody	7 of 17
    7.3       PM25 field QC and maintenance form	13 of 17

    8.1       Cross-sectional view of aPMio sampler inlet head	3 of 6
    8.2       Cross-sectional view of design of a PM2 5 impactor well and filter holder	4 of 6

    9.1       Grounding screw on microbalance	4 of 14
    9.2       Example of environmental logging software displaying both summary
             statistics and atime series graph	5 of 14
    9.3       Second example of environmental logging software display	6 of 14
    9.4       210-Po anti-static strips and gooseneck holder	8 of 14
    9.5       U-electrode (ionizer)  	8 of 14
    9.6       Example calibration certificate for amass reference standard 	11 of 14

    10.1      Filters equilibrating on conditioning trays	4 of 23
    10.2      Relative humidity scenarios 	6 of 23
    10.3      Laboratory relative humidity conditions during a quarter	8 of 23
    10.4      Lab blank control chart	11 of 23
    10.5      Example laboratory data form	13 of 23
    10.6      Filter sample holding time decision flowchart	19 of 23

    11.1      PM2 5 Sample audit data sheet	6 of 15
    11.2      Temperature and humidity verification form	10 of 15
    11.3      PM2.5 Auditing checklist 	12 of 15
                                                 2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. List of Figures and Tables                                       January 2016     Page viii


 List of Tables
 Number                                                                            Page

   2-1       Definitions of Key Terms	2 of 8

   4-1       Acceptance Checks and Limits for Procurement of Equipment and Supplies	3 of 10
   4-2       Tolerance Specifications for Mass Reference Standards	9 of 10

   6-1       Calibration and Verification Check Intervals	17 of 18

   7-1       Sampling Procedure Checks	3 of 17

   8-1       Summary of PM2s Sampler Maintenance Items	2 of 6

   9-1       Filter Preparation and Analysis Checks	13 of 14

   11-1      Systems and Performance Evaluation Components for Manual Method
             PM2 5 Samplers	1 of 15

   12-1      Formulas for Calculations Associated with PM25 Monitoring	1 of 7

   13-1      Summary of Information to be Provided by Sampler	5 of 6

   14-1      Example of Collocated Sampler Requirements	5 of 6
   14-2      Summary of PM2s Number and Type of Collocation	5 of 6
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. Acronyms and Abbreviations
                                               January 2016    Page ix
List of Acronyms and Abbreviations
 AQS
 ANSI
 ASQC
 ASTM
 CFR
 CV
 DMS
 OOP
 DQO
 EPA
 FEM
 FR
 FRM
 HEPA
 NAAQS
 NELAC
 NIST
 NVLAP
 OAQPS
 Pa
 PAMS
 PMio
 PM2.5
 PMP
 210p0
 PQAO
 PSD
 PTFE
 QA/QC
 QA
 QS
 QAPP
 QC
 RH
 SCC
 SLAMS
 SLT
 SOP
 SRM
 SVOC
 Ta
 TSP
 vscc™
 WINS
Air Quality System
American National Standards Institute
American Society for Quality Control
American Society for Testing and Materials
Code of Federal Regulations
coeffi ci ent of vari ati on
data management system
dioctylphthalate
data quality objective
Environmental Protection Agency
Federal Equivalent Method
flow rate
Federal Reference Method
high-efficiency particulate air
National Ambient Air Quality Standards
National Environmental Laboratory Accreditation Conference
National Institute of Standards and Technology
National Voluntary Laboratory Accreditation Program
Office of Air Quality Planning and Standards
pressure, ambient
Photochemical Assessment Monitoring Stations
particulate matter <10 micrometers (um)
particulate matter <2.5 micrometers (um)
polymethylpentene
polonium-210
Primary Quality Assurance Organization
prevention of significant deterioration
polytetrafluoroethylene
quality assurance/quality control
quality assurance
Quality Staff
quality assurance project plan
quality control
relative humidity
sharp cut cyclone
State and local air monitoring stations
State, Local, Tribal monitoring agency
standard operating procedure
Standard reference material
semivolatile organic compound
temperature, ambient
total suspended particulate
very sharp cut cyclone
well impactor ninety-six
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0  Scope and Applicability                                 January 2016    Page 1 of 6


 1.0    Scope and Applicability


 1.1    Overview and Purpose

 This document is intended to provide additional guidance to assist personnel using reference
 methods or Class I equivalent methods to monitor ambient air for particles with an aerodynamic
 diameter nominally equal to or less than 2.5 micrometers (um), known as PIVfo.s. The formal
 specifications for PM2.5 reference method samplers and reference method measurements are set
 forth in the U.S. Environmental Protection Agency's (EPA's) monitoring regulations - 40 Code
 of Federal Regulations (CFR) Part 50, Appendix L (EPA 1997a).* The regulations governing
 the sampling and analytical methods must be followed. This document reviews those formal
 requirements and provides clarifications and supplemental information in greater detail than can
 be provided in the formal regulatory requirements. Because this supplemental information is
 intended for method users, it emphasizes the operational aspects  of the method, rather than the
 design and performance specifications for the sampler. This  document also provides
 recommended quality assurance (QA) procedures and guidance to help monitoring agencies
 reliably achieve the data quality objectives (DQOs) established for PM2.5 monitoring. The goal
 for acceptable measurement uncertainty has been defined as 10 percent coefficient of variation
 (CV) for total precision and ±10 percent for total bias (Papp  et al., 1998). The information
 provided here, together with instruction manuals and technical bulletins provided by the
 manufacturer of the selected PIVb.s sampler, should be used by each monitoring organization to
 develop its own agency-specific quality assurance project plan (QAPP) and standard operating
 procedures  (SOP) to govern its individual PM2.5 monitoring  activity.

 1.2    Applicability

 This document is primarily applicable to PM2.5 ambient air monitoring with reference methods
 carried out by State, Local or Tribal (SLT) air monitoring agencies in their State and local air
 monitoring stations (SLAMS) air surveillance networks under the air monitoring requirements
 of 40 CFR Part 58. The document is also applicable to other organizations required to carry out
 SLAMS or SLAMS-related  PM2.5 monitoring and to any PM2.5 monitoring activity for which
 the PM2.5 reference method measurements will be entered into the Air Quality System (AQS)
 database. The recommendations and guidance in this document—to the extent that they go
 beyond the  specific regulatory requirements set forth in the method or in the sampler-specific
 requirements contained in a particular sampler's operation or instruction manual—are not
 mandatory or binding. In addition to this information, it is suggested that monitoring
 organizations that are implementing gravimetric laboratories consider the requirements and
 guidance in the 2003 National Environmental Laboratory Accreditation Conference (NELAC)
 Standard. Although gravimetric labs are not required to be NELAC certified, following standard
 laboratory procedures described in this document will help ensure the quality and defensibility
 of data. Monitoring agencies are encouraged to consider and adopt these recommendations and
 The 1998 version of this document provided a section for references. Section 15 will continue to include the references from
the 1998 version. Newer references will be included as footnotes to sections and URLs where the information can be found.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0 Scope and Applicability
January 2016    Page 2 of 6
 guidance to help ensure that monitoring data are of acceptable quality. If the monitoring
 organization does not adopt this guidance, it needs to identify and adopt a method considered
 acceptable by the EPA Regions or the QAPP/SOP-approving authority as an acceptable
 alternative to the method and quality assurance requirements described in this document.

 This document is not a full description of a PM2.5  monitoring method and does not substitute for
 the formal method description as set forth in the PIVfo.s reference method (EPA 1997a), or for
 the sampler-specific requirements contained in the associated operation or instruction manual,
 which is an official part of each designated reference or equivalent method for PIVfo.s. Agencies'
 QAPPs and SOPs must meet all applicable Part 50 and 58 requirements for PIVb.s. In addition to
 these requirements, the monitoring organization will be held accountable to the criteria
 documented in their approved QAPPs and SOPs.

 1.3   Conventions

 Where this document refers to mandatory method requirements, the terms "shall" and "must"
 are used, and a supporting reference to the applicable section of the reference method regulation
 or QA requirements is provided. Use  of the word "should" indicates an activity or procedure
 that is recommended to help achieve a high level of measurement data quality but is not
 formally required by the method itself. Finally, the word "may" is used to indicate activities or
 suggestions that are optional or discretionary. The use of these terms is generally consistent
 with American National Standards Institute/American Society for Quality Control
 (ANSI/ASQC)E4-20141 guidance.

 1.4   Format and Structure

 The organization of the document is intended to align more closely with the organizational
 structure and topics of SOPs recommended by EPA's Quality Staff (QS) in its G-6 guidance2.
 This closer alignment will help monitoring agencies use this document to prepare their
 individual SOPs for PM2.5 monitoring. It should be clearly understood, however, that this
 document is not a ready-made, generic SOP and that it needs to be adapted and tailored
 specifically to each monitoring organization's individual policies and circumstances.

 Quality Control (QC) Acceptance Criteria

 The QA acceptance criteria in this document is current as of the date of publication. However,
 as new methods are developed and  data quality evaluations take place in the future, the QC
 requirements may change. This document may not be simultaneously updated to reflect the
 changes that have occurred. However, EPA has developed a validation template for the PIVb.s
 program which contains all the QA/QC criteria found in this document. The validation template
 will be the definitive set of QA requirements and will be regularly updated and posted on the
 EPA's Ambient Monitoring Technology Information Center (AMTIC) website3. Monitoring
1                    search for E-4
2 httJ3://www.ej2a^
3
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0 Scope and Applicability                                 January 2016    Page 3 of 6

 organizations should ensure that the acceptance criteria used is reflective of the requirements
 posted on AMTIC. Each table in this document will provide a footnote to this effect.

 References

 The initial 1998 document included references in the text which were then listed in Section 14.
 While those references remain and continue to be listed in Section 15 of this document, newer
 references are footnoted and URLs provided to direct the reader to a website where the
 referenced information can be found.

 1.5   Overview of Reference and Equivalent Methods for PM2.5

  A "method" for PM2.5 generally consists of the following:

   •   A PM2.5 sampler or analyzer, designed, built and sold by a particular manufacturer

   •   An operation or instruction  manual provided by the instrument manufacturer that
       describes the proper use and operation of the sampler or analyzer

   •   A laboratory that can meet the provisions of Appendix L

   •   The other operational and quality assurance requirements necessary to obtain reliable
       PM2.5 concentration measurements.

 Methods used for monitoring  PM2.5 in SLAMS or SLAMS-related air monitoring networks are
 required to be either reference or equivalent methods, as designated by EPA under the
 requirements and provisions of 40 CFR Part 53. For most SLAMS purposes, reference and
 equivalent methods may be used interchangeably, and a particular method's identity as either a
 reference method or an equivalent method is irrelevant. However, for some purposes, such as
 collocation for QA purposes,  a reference method may be specifically required. A current
 publication, List of Designated Reference and Equivalent Methods, identifying all methods that
 have been designated as reference or equivalent methods by EPA, is available online at the U.S.
 EPA's AMTIC web site4.   It  can also be requested from any EPA Regional Office or by writing
 to Department E (MD-46), U.S. Environmental Protection Agency, Research Triangle Park, NC
 27711.

 1.5.1    Reference Methods

 All formal sampler design and performance requirements and the operational requirements
 applicable to reference methods for PM2.5 are specified in Appendix L of 40 CFR Part 50.
 These requirements are very specific and include explicit design specifications for the type of
 sampler, the type of filter, the sample flow rate, and the construction of the sampler's collecting
 components. However, various designs for the flow rate control system,  the filter holder, the
              ^

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0  Scope and Applicability                                 January 2016    Page 4 of 6

 operator interface controls, and the exterior housing are possible. Hence, various reference
 method samplers from different manufacturers may vary considerably in appearance and
 operation. Also, a reference method may have a single filter capability (single channel sampler)
 or a multi-filter capability (sequential sampler). A PM2.5 method is not a reference method until
 it has been demonstrated to meet all the reference method regulatory requirements and has been
 officially designated by EPA as a reference method for PM2.5.

 1.5.2    Equivalent Methods

 Equivalent methods for PM2.5 have much wider latitude in their design, configuration, and
 operating principles than reference methods. These methods are not required to be based on
 filter collection of PIVh.s; therefore, continuous or semi-continuous analyzers and new types of
 PM2.5 measurement technologies are not precluded as possible equivalent methods. Equivalent
 methods are not necessarily required to meet all the requirements specified for reference
 methods, but they must demonstrate both comparability to reference method measurements and
 similar PIVh.s measurement precision.

 The requirements that some (but not all) candidate methods must meet to be designated  by EPA
 as equivalent methods are specified in 40 CFR Part 53. To minimize the difficulty of meeting
 equivalent method designation requirements, three classes of equivalent methods have been
 established in the 40 CFR Part 53 regulations, based on the degree to which a candidate  method
 deviates from the reference method requirements. All three classes of equivalent methods are
 acceptable for SLAMS or SLAMS-related PM2.5 monitoring, but not all types of equivalent
 methods may be equally suited to specific PM2.5 monitoring requirements or applications.
 Class I

 Class I equivalent methods are very similar to reference methods, with only minor deviations,
 and must meet nearly all of the reference method specifications and requirements.  The require-
 ments for designation as Class I equivalent methods are only slightly more extensive than the
 designation requirements for reference methods. Also, because of their substantial similarity to
 reference methods, Class I equivalent methods operate very much the same as reference
 methods, and most of the information and guidance in this document is applicable to Class I
 equivalent methods.

 Class II

 Class II equivalent methods are filter collection-based methods that differ more substantially
 from the reference method requirements. The requirements for designation as Class II methods
 may be considerably more extensive than for reference or Class I equivalent methods,
 depending on the specific nature of the variance from the reference method requirements. The
 information and guidance in this document may be only partially applicable to Class II
 equivalent methods. The operation or instruction manual associated with Class II equivalent
 methods will, therefore, have to be more detailed in some areas than for reference or Class I
 equivalent methods to provide information and guidance not covered by this document.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0  Scope and Applicability                                 January 2016     Page 5 of 6

 Class III

 Class III equivalent methods cover any PM2.5 methods that cannot qualify as reference or Class
 I or II equivalent methods because of more profound differences from the reference method
 requirements. This class encompasses PM2.5 methods such as continuous or semi-continuous
 PM2.5 analyzers and potential new PM2.5 measurement technologies. The requirements for
 designation as Class III equivalent methods are the most extensive, and, because of the wide
 variety of PM2.5 measurement principles that could be employed for candidate Class III
 equivalent methods, the designation requirements are not explicitly provided in 40 CFR Part 53.
 For similar reasons, much of the information and guidance in this document may not be
 applicable to operation of Class III equivalent methods.

 1.6 Limitations of PM2.s Reference and Class I Equivalent Methods

 There are several conditions or effects that limit the degree to which a PIVh.s reference or Class I
 equivalent method can precisely determine the mass concentration of particulate matter in the
 atmosphere. Procedures to control such effects are discussed throughout this document. One
 limitation is maintaining the specified ambient air flow rate through the sampling inlet and filter
 assembly. The flow rate determines the size of the particles that are collected. The effects of this
 limitation are minimized by following sampler construction requirements and by employing
 procedures and checks to ensure the proper flow rate is maintained within close  tolerances.
 Refer to Section 6.0, "Calibration Procedures," and to Section 7.0, "Field Operation," for
 guidance.

 Another important limitation involves changes in the weight of a collected sample due to
 mishandling, chemical reactions, and volatilization. Handling procedures, choice of filter media,
 humidity and temperature control of the filter (sample) during collection and subsequent
 processing, and promptness in weighing the  sample following collection, all help control filter
 artifacts. The chemical makeup of PM2.5 paniculate matter will vary with sampling location and
 source. Thus, the magnitude of PM2.5 weight changes due to chemical and physical processes
 will also vary with site location.

 Weight loss due to mechanical removal of particles from the filter is minimized  by carefully
 removing the filter cassette assembly from the sampler and storing the filter/cassette in a
 protective container during transit to  the weighing laboratory. Once the filter arrives at the
 laboratory, weight loss can be minimized by carefully removing the filter from the cassette,
 conditioning the filter, and weighing  as soon as the filter conditioning is complete. Refer to
 Section 10.0, "Filter Preparation and Analysis."

 The choice of an essentially non-reactive Teflon® media filter as the collecting surface
 minimizes the weight gain that occurs when sulfate-containing particles form by chemical
 reactions of sulfur dioxide gas at the  surface of alkaline media (such as glass fiber filters).
 Appendix A to this document discusses this effect and includes a procedure for measuring the
 alkalinity of filters. If nitric acid vapor is present at a sampling location, it can deposit on a
 Teflon® filter and cause small weight gains in proportion to the amount of nitric acid present in
 the atmosphere (Lipfert 1994). This weight gain may not be controllable. Weight losses can
 occur due to thermal or chemical decomposition or evaporation of compounds like ammonium

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.1.0  Scope and Applicability                                 January 2016    Page 6 of 6

 nitrate (NFMSTCb), which releases ammonia and nitric acid as gases. Semi-volatile organic
 compounds (SVOCs) may be part of the sample makeup; if so, they may evaporate and cause
 sample weight losses. Such weight losses are minimized or standardized by maintaining the
 sampler filter compartment's temperature near ambient conditions during the sampling process,
 keeping the sample cool during transport to the laboratory, and promptly conditioning and
 weighing the sample following its receipt in the laboratory. Weight gain or loss due to
 absorption or desorption of water vapor on the filter or on the particulate matter is minimized by
 specifying low moisture pickup for manufactured filters and by conditioning the filters within
 specified humidity and temperature ranges, both before use and after receipt from the field.
 Refer to Section 10.0, "Filter Preparation and Analysis," for further guidance.

 Errors in the gravimetric analysis of samples can also result from the buildup of electrostatic
 charges on filters during their manufacture or during sampling  (Engelbrecht et al., 1980).  This
 static buildup will interfere with microbalance weighing, but it can be reduced by using static
 charge reduction techniques. Refer to Section 9.6, "Static Neutralization," for further guidance.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites                                          January2016     Page 1 of8


 2.0    Prerequisites

 2.1    Overview and Definitions

 The air pollutant known as PM2.5 is required by Federal law to be measured and reported on a
 nationwide basis. Regulations governing its measurement were effective September 16, 1997,
 and have been periodically updated.

 This document provides a review of monitoring requirements and offers guidance on the
 collection of samples, the determination of the volume of air sampled, and the gravimetric
 determination of the amount of PM2.5 collected. From the data collected, the concentration of
 PM2.5 particles in the ambient air can be calculated and expressed as micrograms per cubic
 meter (ug/m3) of air. This document addresses procedures that apply mainly to reference and
 Class  I equivalent methods for sampling PM2.5. An annotated listing of acronyms,
 abbreviations, specialized terms, and other expressions is given in Table 2-1 to familiarize the
 new user of PM2.5 samplers with key terms.

 2.2    Personnel Qualifications

 General aspects of personnel  qualifications, training, and guidance are discussed in Section 4 of
 the Quality Assurance Handbook for Air Pollution Measurement Systems Volume //5(EPA
 2013). SLAMS Primary Quality Assurance Organizations (PQAOs) or SLT agencies can use
 information from this source and specific information about PM2.5 sampling and analysis to
 develop a training program for their employees.

 The responsibility for ensuring adequate training of personnel rests with the organization's
 management and should be explained in general in the organization's quality management plan
 (QMP) and specifically in quality assurance project plans (QAPPs). The organization's Quality
 Assurance Coordinator (or equivalent) should be involved as well. He or she should stress to
 management the need for adequate training and recommend that employees be tested or
 examined to evaluate the success of training and identify where further emphasis is needed.

 2.2.1    Laboratory Personnel Qualifications

 All laboratory personnel  should be familiar with clean room environmental laboratory
 procedures and techniques. Those who operate the microbalance in the laboratory need to be
 very conscientious and attentive to details in order to report complete and high-quality PM2.5
 data. Persons involved with PM2.5 laboratory  operations  should be trained and certified to
 perform the following operations:

    •   Use common methods  to determine temperature, pressure, and relative humidity (RH) in
       the laboratory.

    •   Use the microbalance and anti-static devices and information management technology.
 http://www3.cpa.gov/ltn/aiTiiic/qalisi.htiTil

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0 Prerequisites
January 2016    Page 2 of 8
   •   Label, organize, and archive filters and samples in the laboratory.

   •   Use proper filter handling and custody techniques.

   •   Implement all quality assurance requirements, as documented in the regulations,
       applicable QAPPs, and SOPs.
 Table 2-1.  Definitions of Key Terms
Term
Accuracy
Actual conditions
Analyst
American National
Standards Institute
(ANSI)
American Society for
Testing and Materials
(ASTM)
ASTM Class standards
Bias
Cassette
Chain-of-custody
Class I equivalent
method
Class II equivalent
method
Class III equivalent
method
Coefficient of variation
Conditioning
Environment
(weighing room)
Definition
The degree of agreement between an observed value and an accepted
reference value; includes a combination of random error (precision) and
systematic error (bias) components due to sampling and analytical
operations.
The actual ambient temperature and pressure of a gas at the time its
volume (or volumetric flow rate) is measured.
A staff member who weighs the new and used filters and computes the
concentration of PIVh .5 in ug/m3.
Administrator and coordinator of the U.S. private-sector voluntary
standardization system.
A professional organization that develops and distributes protocols for
testing and provides reference standards.
The device of known weight (within some tolerance) for controlling
weighing operations with a microbalance. Certified to ASTM specifications.
The systematic or persistent distortion of a measurement process that
causes errors in one direction.
A device supplied with PM25 samplers to allow a weighed Teflon® filter to
be held in place in the sampler and manipulated before and after sampling
without touching the filter, and to minimize damage to the filter and/or
sample during such activities.
Unbroken trail of accountability that verifies the physical security of
environmental samples and documented information.
An equivalent method for PM25 based on a sampler that closely meets the
specifications for reference method samplers (40 Code of Federal
Regulations [CFR] Part 58, Appendix L), with only minor deviations, as
defined by the U.S. Environmental Protection Agency (EPA).
An equivalent method for PM2 5, as defined by the U.S. EPA, that is based
on a sampler having more significant deviations from reference method
sampler specifications than a Class I equivalent method.
An equivalent method for PM2 5, as defined by EPA, that is based on a
device that incorporates major modifications to the reference method
sampler, or on a fundamentally different design or principle.
A standardized measure of dispersion of a probability distribution or
frequency distribution. It is defined as the ratio of the standard deviation to
the mean.
A confined/defined space containing the microbalance designed to
maintain the PM25 filter conditioning requirements for temperature and
humidity. Filters are held in this area until they have reached a steady state
of moisture.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites
January 2016    Page 3 of 8
Term
Control chart
Conventional mass
Correction Factor
Downtube
Field blank filter
Flow adapter device
Flow calibration device
Flow check filter
Impactor
Impactorwell
Laboratory blank filter
Leak check
Lot Blank
Lot Blank Stability Test
Definition
A graphical presentation of quality control (QC) information over a period of
time. If a procedure is "in control," the results usually fall within established
control limits. The chart is useful in detecting defective performance and
abnormal trends or cycles, which can then be corrected promptly.
Conventional value of the result of weighing in air (see ASTM E617). The
conventional mass of the weight is equal to the assigned nominal weight
plus the correction factor, and will be reported on a certificate of traceability
as the certified weight of the mass reference standard.
As defined in ASTM E61 7, mass values are expressed by two numbers:
one being the nominal mass of the weight, and the second being a
correction. Positive corrections indicate that the weight embodies more
mass than is indicated by the assigned nominal value. Negative corrections
indicate that the weight embodies less mass than is indicated by the
assigned nominal value. The correction, therefore, is equivalent to the error.
The vertically oriented tube that connects the Plvh .5 sampler inlet to sampler
components inside the sampler case. To check the sample flow rate, the
sampler inlet is removed from the downtube and a flow rate standard is
connected in its place.
New filters, selected at random, that are used to determine total
measurement system contamination. They are weighed at the same time
that pre-sampling weights are determined for a set of Plvh .5 filters and used
for quality assurance (QA) purposes.
A tight-fitting connecting device, with an isolating valve, that is inserted in
place of the Plvh .5 sampler inlet on the upper end of a sampler's downtube
and used to connect a flow rate calibration or audit device to check the
sample flow rate. In some cases the device may also be used for leak
checks. Sometimes referred to as the flow shut-off valve.
A National Institute of Standards and Technology (NIST)-traceable flow-
determining apparatus (also called a flow rate standard) that is attached to
the flow adapter device and used to assist in measuring and setting the
volumetric flow rate of air into the sampler. Bubble flowmeters, piston
flowmeters, and orifice devices are the most common types of flow rate
standards.
A standard Plvh 5 filter that is placed in the sampler during the flow check
procedure. It is not used for sampling.
An inertial particle-size separator. A PIVhs reference or Class I equivalent
method sampler uses a specially shaped inlet followed by an impactorthat
allows only particulate matter of well-defined size ranges to penetrate to the
filter collection portion of the collector.
That portion of the sampler inlet where particles larger than 2.5 urn impact
and are held by a filter soaked with oil (tetramethyl tetraphenyltrisiloxane),
such that they are removed from the sample air stream.
New filters, selected at random, that are used to determine laboratory
contamination. The laboratory blank filters shall be weighed along with the
pre-sampling (tare) weighing of each set of PIVhs filters. These laboratory
blank filters should remain in the laboratory in protective containers during
the field sampling and should be reweighed as a quality control check.
A test to determine if any post-air flow is passing the expected flow path of
the PIVhs sampler.
A filter, or group of filters, used to determine filter weight stability over a
period of time due to the volatilization of material from the filter or the
absorption of gaseous material into the filter from the atmosphere.
An annual stability test completed on a lot of newly acquired filters. The test
involves testing 9 random filters (taken from 3 random boxes), and
weighing the filters every 24-hours until each filter shows a measured
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites
January 2016    Page 4 of 8
Term

Mass reference
standard
Microbalance
Method Detection Limit
(MDL)
National Ambient Air
Quality Standard
(NAAQS) for PM2 .5
NIST
Nominal weight
Operator
Orifice flow rate check
device
Pa
PM2.5
PM2 .5 sampler
PM2 .5 Separator
Polonium-210(210Po)
anti-static strip
Precision
Definition
weight change of less than 15ug over a 24-hour period.
ASTM- or NIST-traceable weighing standards, generally in the range of
masses expected for the filters.
A type of analytical balance that can weigh to the nearest 0.001 mg (that is,
1 ug or one-millionth gram).
Lowest concentration that can be detected by an instrument with correction
for the effects of sample matrix and method specific parameters such as
sample preparation.
The NAAQS for PM2.5 is defined in 40 CFR Part 50 as follows:
(a) primary standard: 12 ug/m3, annual mean, based on the 3-year average
of the annual arithmetic mean PM2.5 concentrations; (b) secondary
standard: 15 ug/m3, annual mean, based on the 3-year average of the
annual arithmetic mean PM2.5 concentrations; (c) primary and secondary:
35 ug/m3, 24-hour average, based on the 3-year average at the 98th
percentile of 24-hour PM2.5 concentrations at each population-oriented
monitor within an area.
Acronym for the National Institute of Standards and Technology, which is
the federal technology agency that works with industry to develop and
apply technology, measurements, and standards.
Nominal value represents an accepted condition which is a goal or an
approximation as opposed to the real value, which is always present. Often
a "nominal" value is "de facto" rather than an exact, typical, or average
measurement.
A staff member who services a PM2.5 sampler in routine operation at a field
site and who may also be responsible for transporting the sample filters to
and from the laboratory.
One type of flow rate calibration or check device (transfer standard), often
used in the field, based on an established relationship between flow rate
and pressure drop across the orifice plate. An orifice's operating
characteristics are determined in the laboratory using a flow rate standard
such as a soap film flowmeter. Orifice devices generally require
temperature and pressure corrections.
Local ambient (barometric) pressure.
Particulate matter (suspended in the atmosphere) having an aerodynamic
diameter less than or equal to a nominal 2.5 micrometers (urn), as
measured by a reference method based on 40 CFR Part 50, Appendix L,
and designated in accordance with 40 CFR Part 53.
A sampler used for monitoring PM2.5 in the atmosphere that collects a
sample of particulate matter from the air based on principles of inertial
separation and filtration. The sampler also maintains a constant sample
flow rate and may record the actual flow rate and the total volume sampled.
PM2.5 mass concentration is calculated as the mass of the filter catch
divided by the sampled volume. A sampler cannot calculate PM2.5
concentration directly.
A class of approved devices for removing particles less than 10 urn in
aerodynamic diameter (but greater than 2.5 urn in diameter), but allows
particles of nominally less than 2.5 urn in diameter to pass and collect on a
Teflon® filter surface.
A device containing a small amount of 210Po that emits a particles (He2+)
that neutralize the static charge on filters, making them easier to handle
and their weights more accurate.
A measure of mutual agreement among individual measurements of the
same property, usually under prescribed similar conditions, expressed
generally in terms of standard deviation.
                                                 2.12-PM2.5 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites
January 2016    Page 5 of 8
Term
Primary Standard
Polytetrafluoroethylene
(PTFE)
Qa
QA Collocation
QA supervisor or
coordinator
Readability
Repeatability
Sample Batch
SLAMS
Standard conditions
(EPA)
Ta
Tolerance
Total suspended
particulate (TSP)
Traceable
Transfer standard
Definition
The standard within the lab that is traceable to NIST and has "authority"
over other standards (working standards) in the laboratory; meaning, the
working standards are checked against the primary standard at some
interval or when there is concern about the accuracy of the working
standard.
The polymer that is used to manufacture the 46.2 mm-diameter filters for
PM25 Federal Reference Method (FRM) and Federal Equivalent Method
(FEM) samplers. Also known as Teflon®.
The sampler flow rate expressed at ambient (actual) conditions of
temperature and pressure. Expressed in liters per minute (L/min, or LPM).
The placement of a second PM2 .5 sampler (referred to in CFR as the "QC
monitor") near a permanent network sampler (primary sampler).
Comparison of results from the two samplers is used to estimate the
precision and bias of the primary sampler.
A staff member who assists in preparation of the PQAO's quality plan,
makes recommendations to management on quality issues (including
training), oversees the quality system's control and audit components, and
reports the results.
The smallest difference between two measured values that can be read on
the microbalance display. The term "resolution" is a commonly used
synonym.
A measure of the ability of a microbalance to display the same result in
repetitive weighings of the same mass under the same measurement
conditions. The term "precision" is sometimes used as a synonym.
A grouping of filters that are conditioned and analyzed as a single weighing
event and for which laboratory 24-hour temperature and humidity conditions
are documented and associated with the grouping of filters. The batch
includes the required QC (filter blanks, standards, etc.). A batch may be
based on the number of filters that can be processed in a normal business
day since pre-sampling holding times (30 days) are determined when filters
have been pre-weighed.
Acronym for State or Local Air Monitoring Stations. The SLAMS include the
ambient air quality monitoring sites and monitors that are required by 40
CFR 58, Appendix D, and are needed for the monitoring objectives of
Appendix D, including NAAQS comparisons, but may serve other data
purposes. The SLAMS includes NCore, PAMS, and all other state or locally
operated criteria pollutant monitors, operated in accordance with Part 58,
that have not been designated and approved by the Regional Administrator
as Special Purpose Monitors (SPMs) in an annual monitoring network plan.
EPA-designated atmospheric conditions of 1 atm pressure (760 mmHg)
and 25 °C (298 K). Standard conditions are not used for reporting PM2 .5
data; actual [uncorrected] conditions must be used for reporting.
The ambient temperature. Expressed in degrees Celsius (C).
Also referred to as maximum permissible error (see ASTM E617). It is
defined as the maximum amount by which the sum of the conventional
mass of the weight, its deviation from nominal value, and its associated
uncertainty is allowed to deviate from the assigned nominal value.
Particulate matter collected by a high-volume sampler, which has no
particular particle size selectivity.
A term signifying that a local standard has been compared and certified,
either directly or by not more than one intermediate standard, to a NIST-
certified primary standard, such as a thermometer or standard reference
material (SRM).
A reference standard device (for temperature, flow rate, and so on) that has
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites                                           January 2016     Page 6 of 8
Term

Trip blank
Uncertainty
Va
Very Sharp Cut
Cyclone
WINS
Working Standard
Definition
been compared to a NIST reference standard and is subsequently used to
calibrate, audit, or check other measurement systems or devices.
New filters, selected at random, that are used to determine contamination
occurring during loading filters into cassettes, and/or transportation
activities to/from the field and laboratory.
Spatial, temporal, or measurement variability or bias that leads to a
deviation from the true value.
An air volume measured or expressed at ambient (actual) conditions of
temperature and pressure.
The name given to the design of the particle impactor specified by EPA for
reference method samplers for Plvh .5. Fractionators defined as VSSC and
approved by EPA can be used in place of WINS.
The name given to the design of the particle impactor specified by EPA for
reference method samplers for Plvhs; an acronym for "Well Impactor Ninety
Six."
Standard that is used for calibrations and routine quality control checks. At
some frequency, its accuracy is checked against a primary standard.
 2.2.2   Field Personnel Qualifications

 All field operations personnel should be familiar with environmental field measurement
 techniques. Those who service the PM2.5 sampler in the field need to be very conscientious and
 attentive to detail in order to report complete and high-quality PM2.5 data. Persons qualified to
 perform PM2.5 field operations should be able to:

    •   Operate the PIVb.s sampler.
    •   Perform routine maintenance and cleaning of the PM2.5 sampler.
    •   Calibrate, audit, and troubleshoot the PM2.5 sampler.
    •   Be proficient in the use of microprocessor-controlled instrumentation.
    •   Use common methods to determine temperature, pressure, and flow rate in the field, as
        well as test for leaks.
    •   Enter data into databases, chain-of-custody (COC) forms/labels, and logbooks.

 2.2.3    Training for Laboratory and Field Personnel

 Resources for training laboratory and field personnel should be available. Each individual
 should receive training appropriate to his or her duties on the PM2.5 monitoring program and
 this training should be documented in personnel files. Developing a training manual or
 operations checklist is recommended.  Suggestions for training activities/materials include:

    •  Reading and understanding the pertinent PM2.5 section of CFR.

    •  Studying the relevant sections of this QA guidance document, the operating manuals for
      the samplers and/or the microbalance, and other guidance.

    •  Attending relevant EPA or SLT sponsored training workshops.

    •  Watching and studying instructional videos that may become available.

                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites                                          January 2016    Page 7 of 8

   •  Having experienced operators instruct others through on-the-job training and attest to the
      readiness of new operators.

   •  Having newly trained operators pass a hands-on test of procedures and attest that they
      have read and understand the program procedures and requirements.

 2.3    Health and Safety Warnings

 To prevent personal injury, all employees should heed any warnings that are associated with
 operation of the microbalance, the PIVb.s sampler, and any supporting equipment and supplies.
 Specific health and safety warnings are generally found at the point in the operating manual or
 troubleshooting guide where they are most applicable.

 In general, health and safety warnings fall into these categories:

    •  Electrical
    •  Chemical
    •  Equipment placement and stability.

 Electrical safety considerations that apply to the PIVh.s sampler include the following:

  •  Make all electrical connections in accordance with national codes. Use of a third-wire
     grounding arrangement, a ground fault circuit interrupter (GFCI) outlet and cord on
     samplers, electrical appliances or test rigs, will minimize the possibility of electrical shock
     and injury.

  •  Electrical supply lines to the PM2.5  sampler should be installed so that they are protected
     from degradation and hazards. Inspect the electrical cords and connections for signs of
     wear and have an electrician repair or replace them as needed.

  •  The power to the sampler should be unplugged when servicing or replacing parts in areas
     requiring removal of protective panels.

  •  Avoid contact of jewelry with electrical circuits. Remove rings, watches, bracelets, and
     necklaces to prevent shorting and electrical burns.

  •  Use caution when working near moving parts (such as pumps) to avoid injury.

  •  Be aware of weather patterns; avoid the sampling platform if lightning storms approach
     because the sampler is elevated and made of conductive metal.

 Chemical and physical safety considerations that apply to PM2.5 operations include the
 following:

   •   Use care in the application of cleaning solvents, greases for O-rings, and silicone oil for
       the impactor well. Use of gloves is recommended. Wash hands thoroughly  after working

                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.2.0  Prerequisites                                           January 2016    Page 8 of 8

       with chemicals. Provide good ventilation if organic solvents are used. Dispose of
       chemicals and shop towels properly.

   •   Mercury metal,  a poisonous material, is present in some types of barometers and RH
       indicators. In most cases, these instruments have been replaced in the field with non-
       mercury containing devices. If liquid mercury is spilled, it must be cleaned up and
       disposed of properly. Use protective equipment to avoid inhalation of vapors and
       impermeable gloves to avoid skin contact. Mercury cleanup kits are available. Avoid use
       of mercury, if possible, to preclude mercury exposure.

   •   Exercise caution when using anti-static devices containing radioactive polonium sources.
       Keep an inventory of the location and size of anti-static devices. Dispose of the devices in
       accordance with State and local regulations.

   •   Avoid touching pump or fan motors, which may be hot. Avoid injury from moving parts,
       such as fans  and filter-changing mechanisms.

   •   Ensure that the PM2.5 sampler(s) and associated equipment are stably mounted to the
       ground or another surface so that they do not tip over.

   •   Electrical wires should be installed so no one will trip over them and be injured.

   •   If the sampler is placed on the roof of a building or on another elevated location, railings
       should be installed and warning signs placed to prevent falls.

 2.4    Cautions

 Damage to the PM2.5 sampler may result if caution is not taken to properly install and maintain
 the device. Consider the manufacturer's instructions for maintaining the pump, for  cleaning the
 interior and exterior surfaces, and for safe, secure installation.

 The weighing room should be maintained in a state of good order so that samples are not con-
 taminated, misplaced,  or misidentified. The room air should be dust-free and controlled within
 specified temperature and humidity ranges. The microbalance should be placed on  a vibration-
 free mounting. Care must be taken to avoid bumping the balance and disturbing its calibration
 settings. The calibration weights used to check balance precision and accuracy should be stored
 in a secure location away from sources of corrosion and should be used only for filter weighing
 by a knowledgeable analyst and handled only with Teflon® forceps to prevent scoring and
 weight changes.

 Finally, exercise care in handling new and used filters. If details concerning weighing, labeling,
 and transporting filters are not followed precisely, errors may result. Rough handling of used
 filters may dislodge PM2.5 material. Inadequate conditioning of filters at the specified
 temperatures and humidities or long delays between sample retrieval and sample weighing may
 lead to positive or negative weight changes and, thus, inaccurate PM2.5 concentrations.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.3.0 Summary                                              January2016    Page1of5


 3.0    Summary

 3.1    Overview

 The procedures in this document are designed to serve as guidelines for the development of
 SLT agency quality assurance (QA) programs in support of measurements of PM2.5 using either
 Federal Reference Method (FRM) samplers or Class I equivalent method samplers. Section 3.1
 gives background information on the development and intended uses of FRM and Class I
 equivalent method samplers; Section 3.2 illustrates the design of the PM2.5 sampler and
 describes how PM2.5 particles are captured; Section 3.4 contains a checklist of the typical steps
 involved in servicing a sampler in the field.

 As described in 40 CFR Part 50, Appendix L, the reference method for PM2.5 sampling employs
 a sampler to draw a measured quantity of ambient air at a constant volumetric flow rate (16.67
 L/min) through a specially designed particle-size discrimination inlet. PM2.5 particles are those
 particles with an aerodynamic diameter less than or equal to a nominal 2.5 um. Particles in the
 2.5 um and smaller size range are collected on a 46.2-mm- diameter Teflon® filter during the
 specified 23- to 25-hour sampling period. Each filter is weighed before use and after sampling.
 From these measurements, the mass of the collected PM2.5  sample can be calculated.

 Filter samples are required to be collected on daily, l-in-3 day, and l-in-6 day sampling
 schedules. The national sampling schedule is posted on AMTIC6. The samplers are operated
 midnight-to-midnight, local standard time.

 The total volume of air sampled is determined from the measured volumetric flow rate and the
 sampling time. The mass concentration of PM2.5 in the ambient air is computed as the total mass
 of collected particles in the PM2.5 size range divided by the total volume of air sampled and
 measured under ambient (actual) conditions of temperature and pressure. The PM2.5 concentra-
 tion is thus expressed as ug/m3 of air. The particle-size discrimination characteristics of the
 sampler inlet, downtube, separator, and filter holder are all specified by design in 40 CFR Part
 50, Appendix L. In addition, specific performance characteristics of the sampler are tested in
 accordance with the procedures in 40 CFR Part 53, Subpart E. Sampling methods for PM2.5 that
 meet all requirements in both Parts 50 and 53 are designated as PM2.5 FRMs for use in SLAMS
 and Prevention of Significant Deterioration (PSD) monitoring networks. These designated
 methods are identified by a specific number and can also be identified by the manufacturer and
 model number of the sampler.
6 http://www.cpa.gov/ttn/amlic/calcndar.html

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 3.0 Summary
                                                               January 2016     Page 2 of 5
 3.2    Illustrations of Sampler Inlet Components and Sample Flow Paths
                                                    Ambient Air Flow
Figure 3-1 illustrates the inlet of the PM2.5
sampler. This inlet is designed to representa-
tively extract ambient aerosols from the
surrounding airstream, remove particles with
aerodynamic diameters greater than nominally
10 um, and send the remaining smaller particles
to the next stage. Figure 3-2 illustrates the
impactor and filter holder assembly that first
removes those particles nominally less than 10
um but greater than nominally 2.5 um in
diameter, but allows particles of nominally 2.5
um in diameter (and smaller) to pass and collect
on a Teflon® filter surface. This separator has
been called the "Well Impactor 96", or WINS
impactor. Downstream of the inlet, particles
less than 10 um but greater than 2.5 um are
removed by a PM2.5 separator. The well of the
impactor assembly contains a 37-mm-diameter
glass fiber filter immersed in 1 mL of low-
volatility, low-viscosity diffusion oil. The oiled
glass fiber filter removes particles between 10
and 2.5 um in  diameter by preventing bouncing of the incoming particles. An alternative to
Figure 3.2 is the very  sharp cut cyclone (VSCC), illustrated in Figure 3.3, that has been
approved by EPA through the FRM/FEM designation process.
                                                               Flow to PM2.5 separator
                                                Figure 3.1 PM2 5 inlet
                 From Aerosol Inlet
  Fine Partides, <2.5 (jrn
                              Impactor Nozzle
                               Irnpaclor Well
                               Coarse Partides, >2 5 pm
                              Finer Cassette
                                                              ABOSaFROIMPM.FBMIN.ET a 16
                                                                 FW.. A£ROSCt TO OlTEft
  Figure 3.2 Air flow through impactor well and
  filter holder
                                                Figure 3.3 Example of a very sharp cut cyclone
   NOTE: Method designations, which certify that a particular FRM/FEM sampler or analyzer
   has been formally approved by EPA, are partially based on the type of size-selective
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 3.0 Summary                                              January 2016     Page 3 of 5

   separator used in the instrument. In most cases, the last three digit value in the method
   designation is used to identify the "method code" of the sampler in AQS.   This may not
   always be the case. Samplers may be designated for use with either VSCC™ or WINS and
   have only one method designation. However, in AQS they will be required to have a separate
   method code to identify the separator being used. Use of a different separator than what was
   originally approved for a method designation is not permitted and may change whether the
   sampler is designated as FRM or FEM. Take care to check on the correct method designation
   and approved separator before changing the type of separator used in a designated
   FRM/FEM sampler. In addition, some separators may not be approved as FRM/FEM for
   some samplers. Therefore, purchasing of less expensive separators that have not been
   designated for samplers may void the FRM/FEM designation. Additional information
   regarding approved particle-size separators is updated semi-annually in the List of
   Designated Reference and Equivalent Methods, available on AMTIC7.

 3.3    Summary of Field and Laboratory QC Checks

 The validation templates provided in the QA Handbook Volume II, and posted separately on
 AMTIC8, summarize the field and laboratory QC checks required to collect and process PM2.5
 samples in accordance with U.S. EPA regulations. The frequencies of the checks are listed and
 cross-references are made to the Code of Federal Regulations and to this guidance document.
 Since the QC requirements occasionally change, and EPA cannot always revise documents in a
 timely manner, it is more convenient to have these tables available as separate document that
 can be updated more frequently. The validation templates, therefore, are not included in this
 document.

 3.4    Summary of Sampling Procedures for PIVk.s

 To illustrate the steps to install a filter/cassette in the sampler to begin a run, as well as the  steps
 to shut down the sampler at the end of a run,  an outline is provided as an example procedure.

 I. Prepare for Site Visit on Scheduled Date/Time
 Be sure to  take the following equipment and supplies to each of the sites:

   •  Pre-weighed sampling filter(s) in cassette(s), packed in  labeled container(s). Spares.
   •  Pre-weighed field blank filter(s) in cassette(s), packed in labeled container(s), if a field
      blank study is scheduled.
   •  PM2.5 sampler run data sheet for each sampler. Use of electronic data sheets is encouraged.
      Add  comments to the data sheets. Site notebook and calculator.
   •  Laptop computer (and connecting cables)  or alternate data storage device, if applicable, to
      download sampler data.  NOTE: Some agencies may have developed electronic access
      capability with their samplers and, therefore, may be able to directly  download/upload
      data remotely.
  •   Spare parts and tools including O-rings, silicone grease, laboratory wipes, voltmeter.
7 http://www3.epa.gov/ttnaJTitil/criteria.htiTil
8 http://www.epa.gov/ttnajTitil/files/aiTibient/piTi25/qa/appd validation template_amtic.pdf
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 3.0 Summary                                               January 2016     Page 4 of 5

   •   Operator's manual for the sampler(s) to be serviced.
   •   If it is time for the recommended separator maintenance, also take the following:
          -  Ready-to-use impactor well assembly or filter/lab wipes/diffusion oil to service
              the sampler at the site, or
          —  A secondary separator or equipment for cleaning the secondary separator in the
              field.

II. Install Filter/Cassette and Begin Sampler Operations
     NOTE: If a filter is still in the sampler, follow the instructions for removing the
     sample/cassette. Also, if the schedule calls for any of the following procedures,
     perform them before installing a new sampling filter/cassette:

     -   Impactor well or VSCC replacement or cleaning (after every fifth sampling
         event);
     —   Sampler flow verification check (every 4 weeks);
     -   Sampler ambient pressure and temperature checks (every 4 weeks);
     —   Sampler verification/calibration (temperature, pressure, flow rate);
     —   Cleaning of sampler inlet head and downtube (monthly);
     —   Leak check before each flow rate verification and calibration;
     -   Leak check before and after separator maintenance;
     —   Field blanks.
Perform the following steps to install the filter/cassette and begin sampler operations:
   •   Be sure the sampler is not in sampling mode.
   •   Fill in initial information on the data sheet.
   •   Remove the new filter/cassette from its protective container and visually inspect the
       filter/cassette for flaws. Verify that this is the correct filter for this sampler, site, and
       run date.
   •   Remove the sampler's filter holder assembly (if required by the manufacturer's
       instructions). Inspect the O-rings inside the filter holder; service them if necessary.
   •   Install the filter/cassette in the filter holder assembly and then install the loaded filter
       holder assembly in the sampler based on the manufacturer's instructions. If you touch
       or scratch the filter, void the filter, record this fact, and get another one from the set of
       extra filters brought to the site.
   •   Program the sampler to turn on at the beginning of a sampling period (consult the
       sampler's instruction manual). The sampling day begins at midnight.

III. Remove Filter/Cassette and End Sampler Operations
Perform the following steps to remove the filter/cassette and end sampler operations:
   •   Arrive at the site as soon as possible after the end of the sampling period. The
       filter/cassette must be removed within 177 hours after the collection period has ended.
   •   Review the sampler data. Record the stop time; total elapsed time; initial and final Q,
       Qavg, and Qcv; total volume sampled; Ta; Pa; and so forth on the data sheet.

                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 3.0 Summary                                               January 2016     Page 5 of 5

   •   For each completed run, download data from the sampler data port to a laptop or
       alternative data storage device.
   •   Open the filter holder assembly (consult the sampler's instruction manual); remove the
       used filter/cassette; quickly visually inspect the filter for tears, oil, insects, moisture, and
       so on; and record your observations on the data sheet.
   •   Place the filter/cassette inside a properly labeled protective container. Verify the
       container's label versus the site name, date, and  so on.
   •   Place the filter/cassette in its protective container inside a cooled storage chest. Do not
       allow the protective container to come into contact with ice or water. Sealed cooling
       blocks are recommended. Protect the containers from condensed water.
   •   Inspect the interior of the sampler filter housing. Note any abnormalities.
   •   Inspect the interior of the impactor housing and the exterior of the impactor well or
       secondary separator. Remove any moisture or dust with a lint-free wipe and make notes
       on the data sheet.
   •   If another sampling run is to begin, insert a new filter/cassette in the filter holder
       assembly and set up the sampler for the next run.
   •   Interpret the recorded data for start and end times, sample elapsed time, flow rate, filter
       quality,  and temperature to start the process of determining if the  sample is valid,
       questionable, or invalid. Scan through the sampling summary on the  sampler display and
       note flags. Record observations and reasoning for questioning or invalidating a run on the
       data sheet.
   •   Make a  final check of the site, and observe and record the presence of any activity that
       may have affected the particulate loading of the  sample.
   •   Keep the container holding the filter/cassette at a temperature below  25°C (preferably
       cooled to 4°C), but not frozen. Promptly deliver or ship the container and the data sheet
       (with comments) to the sample custodian  or balance operator in the weighing laboratory.
       Or, if there is a delay in delivering/shipping the sample to the laboratory, place it in a
       refrigerator until such time as it can be delivered/shipped. Do not place samples in a
       freezer!
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                   January 2016   Page 1 of 10


4.0   Procurement of Equipment and Supplies

 4.1    Overview

 The establishment of an ambient PM2.5 air monitoring network requires the procurement of
 specialized equipment and supplies for field operations and subsequent laboratory determination
 of particle mass. Information in this section will assist PQAOs and SLTs in selecting the proper
 equipment and in acceptance testing that equipment. Section 4.2 describes field operation
 equipment (excluding sampling filters) and Section 4.3 describes laboratory equipment
 (including sampling filters). Note that sampling filters are described in Section 4.3 because they
 are purchased for the laboratory, which must equilibrate and weigh them before field use. In
 addition to field operations and laboratory equipment, data handling system(s) (including forms,
 logs, files, and reporting procedures) should be developed and implemented. Information on
 data handling systems is provided in Section 13.0, "Data and Records Management."

 Minimum monitoring equipment requirements and budgetary limits should be established
 before equipment is purchased. In addition, acceptance criteria for equipment and supplies
 should be established. Upon receipt of the sampling equipment and supplies, acceptance checks
 should be conducted. The results of these checks, including whether the equipment was
 accepted or rejected, should be recorded in a procurement log. Figure 4.1 is an example of such
 a log. This log will  serve as a permanent record for procurement and provide cost projections
 for future programs. Table 4-1 lists the major equipment needed, how it should be tested,
 suggested acceptance limits, and actions to be taken if acceptance limits are not met.

 It is highly recommended that agencies develop an equipment replacement plan that covers
 parts/equipment needed for the next 5 years. It is common practice, as technology advances, for
 vendors to develop  better samplers/calibrators  and discontinue supporting the older models.
 Agencies should plan for replacement of samplers and transitioning to better methodologies as
 they are proven and become available.

 It is also highly recommended for agencies operating their own weighing laboratories to have a
 contingency plan for weighing filters. Good laboratory environmental control and stability are
 two of the most important components of filter weighing and should not be overlooked. If there
 is a problem in the laboratory and filter weighing cannot occur, delays in weighing can put
 samples at risk of invalidation. Many of the validation criteria for PM2.5 filter weighing are time
 sensitive, and a laboratory cannot afford down-time. For these reasons,  a laboratory should have
 a contract or an agreement with another PM2.5 weighing laboratory in the event of a lab issue to
 continue filter weighing.

 It is impossible to include in this document every item or service that might be needed by any
 specific monitoring network. Therefore, each monitoring organization should determine the
 extent of its in-house inventory or capability and the items that should be procured before
 sampling begins. The monitoring organization should also be prepared to order any additional
 equipment or  services required in addition to that outlined in this section.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies
January 2016   Page 2 of 10

Item

1 case
filters







2 |jm
pore
46.2 mm
diameter



Qty

60






PO #

971-100






Vendor

WIZ
Supply




Date

Ord.
8/1/97






Rec'd.
8/15/97






Cost

$100






Initials

ABC





Accept/
Reject

Accept






Comments

Examined
8/20/97




 Figure 4.1. Procurement Log

 4.2    Procurement Prerequisites—Field Operations

 4.2.1    Reference or Equivalent Method Sampler

 Each PM2.5 sampler used for reporting data to determine attainment of the National Ambient Air
 Quality Standard (NAAQS) for fine (2.5 um) particulate matter must meet U.S. EPA standards
 and must be of a model designated by EPA as a reference or equivalent method. The minimum
 sampler requirements are described in the reference method (EPA 1997'a) and in 40 CFR
 Part 53 (EPA 1997b). The PM2.5 sampler cost will vary with the manufacturer and the
 sophistication of the sampler. Basic considerations for sampler selection include flow control
 and measurement systems, maintenance requirements, reliability, ease of operation, and such
 additional capabilities as  sequential sampling and downloading data.

 Although a network may  decide to use any number of different reference and designated
 equivalent PM2.5 sampler models,  using a single model of sampler in a network minimizes the
 variety of spare parts required to keep the network in operation, simplifies training of operators,
 and adds consistency to the data reduction process. An in-house inventory of general
 maintenance supplies and replacement parts is recommended. Examples include: various hand
 tools, laboratory wipes, and soft brushes. Spare parts for the sampler may be obtained from the
 manufacturer or may be purchased from other suppliers.

 In accordance with 40 CFR Part 50, Appendix L, Section 7.3.4.2, the impactor in the WINS
 PM2.5 separator (see Figure 3.2) requires a circular, 35- to 37-mm-diameter filter made  of
 borosilicate glass with no binder. The filter pore size must be 1 to 1.5 um and the thickness
 must be 300 to 500 um. The impactor oil is tetramethyltetraphenyl-trisiloxane, single-compound
 diffusion  oil, with the following specifications at 25 °C: vapor pressure, 2 x 10"8 mmHg;
 viscosity, 36 to 40 centistokes; and density, 1.06 to 1.07 g/cm3 (see Appendix L, Section
 7.3.4.3). In November 2000,  a second type of oil, identified as dioctyl sebacate (DOS), was
 approved for use and is now recommended for use in place of
 Tetramethyltetraphenyltrisiloxane 9.
9 http://www3.epa.gov/ttn/amtic/pmpolgud.html
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies
January 2016   Page 3 of 10
         Table 4-1. Acceptance Checks and Limits for Procurement of Equipment
                                       and Supplies
Equipment
Field operations
Sampler
Calibration quality
assurance/quality control
(QA/QC) equipment for
flow rate, temperature,
pressure, leaks, time
Audit equipment
Acceptance check

Sampler and accessories
complete; no evidence of damage.
Model designated as reference or
equivalent method. Sampling
system verification.
Accompanied by certificate. Check
values against National Institute of
Standards and Technology (NIST)-
traceable standards.
Same as for calibration equipment
but must not be the same
equipment.
Laboratory operations
Filters, Teflon®
Filter cassettes
Filter/cassette
protective containers
Filter-handling
containers
Analytical
microbalance
Mass reference
standards
Temperature and
Relative Humidity
Logger
Audit equipment
Of correct type and undamaged.
Of correct type and undamaged.
Of correct type and undamaged.
Of correct type and undamaged.
Accompanied by certificate; check
values against working standards.
All standards accompanied by
NIST-traceability certificates.
Accompanied by certificate. Check
values against National Institute of
Standards and Technology (NIST)-
traceable standards. Data
collection software validated.
Same as for calibration equipment
but must not be the same
equipment.
Acceptance limits10

Specifications outlined in
40 Code of Federal
Regulation (CFR) Part
50, Appendix L
Within accuracy limits
described in this
document
Within accuracy limits
described in this
document1

Type as described in 40
CFR Part 50, Appendix L
As specified by sampler
manufacturer
As described in this
document
As described in this
document
Readability 1 (ig,
repeatability 1 jig
±25 |ig tolerance or
better (see Table 4-2)
±1.0 °C accuracy for
temperature
±2% RH accuracy for
relative humidity
Capable of 5-minute
averaging/logging
Within accuracy limits
described in this
document1
Action if require-
ments are not met

Reject sampler
Adjust or reject
equipment
Adjust or reject
equipment

Reject filters
Reject filter
cassettes
Reject protective
containers
Reject filter-
handling containers
Adjust or reject
equipment
Reject standards
Reject device
Adjust or reject
equipment
1 audit standards must be the same or of higher accuracy than working or field transfer standards
10 See Validation Template on AMTIC for any revisions to acceptance criteria in this table
http://www3.cpa.gov/ttn/amtic/qapolli.itant.html
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                   January 2016  Page 4 of 10

 4.2.2    Calibration Equipment

 Specialized equipment is needed to calibrate the sampler for temperature, barometric pressure,
 and volumetric flow measurements. At a minimum, the following equipment should be
 procured:

 •  A thermometer(s) capable of measuring ambient temperatures in a range of -30 to +45 °C,
    readable to the nearest 0.1 °C. This thermometer should be referenced (ASTM 1992; EPA
     1995; NIST 1986, 1988, 1989) to within an accuracy of ±0.5 °C to NIST-traceable
    American Society for Testing and Materials (ASTM) precision thermometers (e.g., ASTM
    numbers 62C, 63C, and 64C) once a year (ASTM 1995a). Multiple thermometers may be
    used to cover the temperature range as long as each thermometer meets the accuracy and
    readability specifications described above. Thermistor or thermocouple thermometers for
    use in the field should be calibrated against NIST-traceable thermometers. Some flexibility
    in these recommendations applies at sites where temperatures are extremely cold.

 •  A barometer capable of measuring barometric pressure (ASTM 1995b) over a range of 600
    to 800 mm Hg (80 to 106 kiloPascals [kPa]) and readable to the nearest  1 mmHg.  At least
    once a year, this barometer should be calibrated to within ±5 mmHg of a NIST-traceable
    barometer of known accuracy. A portable, aneroid barometer (e.g., a climber's or  engineer's
    altimeter) is suitable for field use. A Fortin-type, mercury-column barometer may also be
    useful for laboratory checks of the barometer used for field measurements.

 •  Flow rate measurement equipment (transfer standards) must be capable of calibrating or
    verifying the sampler's flow rate measurement device within an accuracy of ±2 percent.
    This flow rate standard should be a separate, stand-alone  device. It must have its own
    certification and be traceable to a NIST primary standard for volume or flow rate.  A variety
    of flow rate transfer standards, with their optimum flow ranges and their support equipment,
    are presented in Section 6.3.2.

 •  A flow adapter of the dimensions specified in drawing No. L-30 in Appendix L of 40 CFR
    Part 50 (EPA 1997a) and described in Section 7.3.6, Appendix L, that will connect the
    transfer standard outlet to the PM2.5 sampler flow path entering the impactor assembly and
    form a leak-free seal. All interconnecting tubing should be flexible and crimp-resistant. This
    adapter is also used in conducting the required external leak test.

 •  A timepiece or device capable of accessing the NIST atomic clock to the second.

 4.2.3    Flow Verification Check Device

 As part of the quality control (QC) procedures, a flow verification check device is required to
 verify that the PM2.5 sampler is operating at the correct flow rate. A variety of flow check
 devices are  currently available on the market. The monitoring organization should select a flow
 device with an operational range that includes the targeted flow rate of the PM2.5 sampler
 (-16.67 L/min, or LPM). Current flow rate measuring devices operate using a measuring device
 of some kind (bubble, piston, and orifice) integrated with a microprocessor. These devices are
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                    January 2016   Page 5 of 10

 not primary standards and require a periodic calibration or certification, typically annual,
 against a NIST standard from a metrology laboratory to demonstrate NIST-traceability. All of
 these standards require time to equilibrate to local ambient conditions before use. These devices
 are typically very user-friendly and designed for work in the field or laboratory. Although not
 required, separate flow rate calibrations devices and verification devices are encouraged. Since
 flow rate audits only occur every 6 months (flow rate audit devices must be separate from
 verification devices), it may take a considerable amount of time to identify unacceptable
 systemic errors which can lead to significant data invalidation if flows rate are calibrated and
 then verified with the same faulty flow rate standard. Detailed descriptions of different types of
 volumetric flow rate devices may be found in Section 6.3.2.

 4.2.4    Audit Equipment

 Audit equipment should be similar to the calibration equipment described above. However, the
 audit equipment must not be the same equipment used for calibration and flow verification
 checks, although both sets of equipment may be of identical makes and models (see 40 CFR
 Part 58, Appendix A, Section 3.2.2). The calibration equipment and the audit equipment may be
 referenced to the same NIST primary standard.

 4.3    Procurement Prerequisites—Laboratory Operations

 4.3.1    Filter Media

In the laboratory,  sample filters are conditioned (while in glass or plastic filter-handling
containers,  such as petri dishes or petri slides), pre-weighed, and placed first into sample filter
cassettes and then into protective containers for transport to the field. The mean relative humidity
(RH) in the filter conditioning environment must be held between 30 and 40 percent, with a
variability of not more than ±5 percent RH over 24 hours. However, where it can be shown that
the  mean ambient RH during sampling is less than 30 percent, conditioning is permissible at a
mean RH within ±5 percent RH of the mean ambient RH, but in no case at an RH less than 20
percent. Mean laboratory temperature must be held between 20 and 23 °C, with a variability of
not more than ±2 °C over 24 hours. The time required for the weights of new filters to stabilize
(see Section 10.4) may be as long as 6 weeks to eliminate initial outgassing. Adequate numbers
of filters should be purchased to permit conditioning before use. In practice, this suggestion
means that  networks should plan on keeping a minimum filter inventory sufficient for 6 to 12
weeks of operational needs beyond the annual order. These inventory levels can be adjusted
based on the network's experience with filter weight stabilization times for a given vendor's
filters.  The EPA filter procurement vendor will precondition the filters before shipping to reduce
the  effects of off-gassing during the production process. This action should lessen the pre-use
conditioning time significantly. EPA-purchased Teflon® filters will be acceptance-tested prior to
distribution to SLT agencies. However, filters should be inspected for obvious flaws (see Section
10.3). Conditioning for outgassing will still be necessary to determine the time, in excess of the
24 hours, needed to obtain a stable weight.

 Only sampling filter media that meet the following EPA specifications for use with PM2.5
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                   January 2016   Page 6 of 10

 samplers should be purchased. The manufacturer must certify that these media meet the
 specifications given in Appendix L of 40 CFR Part 50 (EPA 1997a).

 •    Size—circular, 46.2-mm diameter ±0.25 mm (with support ring)

 •    Medium—polytetrafluoroethylene (PTFE) Teflon® with integral support ring

 •    Support ring—polymethylpentene (PMP) or equivalent inert material, 0.38 ± 0.04 mm
      thickness, outer diameter 46.2 ± 0.25 mm, and width of 3.68 mm. The support ring
      diameter should also be checked for consistency. Variations in diameter will affect the
      filter exposure area, which, in turn, will affect any operations that involve use of a portion
      of the filter.

 •    Pore size—2 um as measured by ASTM F 316-94

 •    Thickness—30 to 50 um

 •    Maximum pressure drop (clean filter)—30 cm FhO column at 16.67 L/min clean air flow

 •    Maximum moisture pickup—No more than 10 ug weight increase after a 24-hour
      exposure to air at 40 percent RH, relative to the weight after a 24-hour exposure to air at
      35 percent RH

 •    Collection efficiency—greater than 99.7 percent, as measured by the dioctyl phthalate
      (DOP) test (ASTM 1995c) with 0.3-um particles at the sampler's operating face velocity

 •    Filter weight stability (including test for loose, surface-particle contamination and test for
      temperature stability)—filter weight loss <20 ug in either test, measured as specified in
      40 CFR Part 50, Appendix L, Section 6.9

 •    Alkalinity—less than 25 microequivalents/g of filter, as measured in a procedure based on
      the one given in Appendix A to this document.

 Although not required for determination of PM2.5 mass concentration under this reference
 method, additional specifications for the filter should be developed by users who intend to
 subject archived PM2.5 filter samples to subsequent chemical analysis. These supplemental
 specifications include background chemical contamination of the filter and any other filter
 parameters that may be required by the chemical analysis method. All such supplemental filter
 specifications should be compatible with and secondary to the primary filter specifications
 given in Appendix L of 40 CFR Part 50 (EPA 1997a).

     NOTE: EPA implements a national filter contract that support SLTs agencies for all PM2.5
     SLAMS filter needs. On an annual basis, SLTs will request their filter needs to their
     respective EPA Regions, who will provide those orders to the EPA Office of Air Quality
     Planning and Standards (OAQPS).
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies
                                                       January 2016   Page 7 of 10
 4.3.2     Filter Support Cassettes

 Most PM2.5 samplers use filter support cassettes (see 40 CFR Part 50, Appendix L) to hold the
 sampling filter while in use. Cassette designs may vary slightly for different model samplers
 and so may not be interchangeable from one sampler model to another. They should be
 purchased only from the sampler manufacturer or other source approved by the sampler
 manufacturer. In either instance, the filter support cassettes must meet the design specifications
 referenced in 40 CFR Part 50, Appendix L, Figures L-27 through L-29. The filters are placed
 into these cassettes in the laboratory. Filter support cassettes should be uniquely identified. The
 cassettes, with the filters in place, are then transported to the field in protective containers (see
 Section 4.3.3). A sufficient number of cassettes should be purchased to allow all placement and
 removal of filters in cassettes to be performed in the laboratory. Under no circumstances should
 filters be placed in or removed from the filter support cassettes in the field.
 4.3.3     Filter Cassette Protective Containers

 Filter cassettes should be transported between the laboratory and the field in protective
 containers, as illustrated in Figure 4.2. Suitable containers should be available from the
 manufacturer or from a supplier specified by the manufacturer. Containers should be designed
 to protect the cassette and prevent it from moving within the container during transport.
 Containers should be fabricated from materials that have been demonstrated not to cause a static
 charge buildup. Static-proof bags are also an option. To avoid losses of particulate matter, do
 not allow the surface of the filter to touch any interior surfaces of the protective container.
 Containers should be marked for identification  purposes. Special, sampler-specific protective
 containers are needed for the trays or cartridges that hold multiple filters for use in sequential
 samplers. Consult the sampler or cassette manufacturer for details and availability. If reuse of
 the containers is desired, a cleaning procedure that does not cause further contamination or
 degradation of the container should be developed.
                                               Anti-static
                                               Re-sealable
                                                  Bag
Metal Canister    Metal Cassette
                     Caps
 Figure 4.2 Example equipment for transporting filter cassettes
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                    January 2016   Page 8 of 10

 4.3.4    Insulated Shipping Container

 An insulated shipping container, capable of maintaining a temperature below 25 °C, is needed
 to transport the filter cassette containers between the field and laboratory after sampling (see 40
 CFR 50, Appendix L, Section 10.13). Plastic coolers, such as those used by campers and
 fishers, filled with leak-proof ice substitutes have proven suitable for this use.  The sample
 cassette temperature should be monitored from the time of collection until the  sample is
 equilibrated (i.e., during shipping, handling, and storage). Monitoring, during filter transport,
 can be accomplished by placing an inexpensive electronic max-min thermometer in the shipping
 container for each leg of the trip to the laboratory. Other devices, such as thermologgers, are
 available for use as well.

 4.3.5    Filter Handling Containers

 Filter media (particularly the 46.2-mm-diameter [1.85-in], 2-um pore-size filter specified for
 use with reference or Class I equivalent samplers) are especially delicate and easily damaged.
 Post-sampling particle loss and filter damage may occur if proper handling procedures are not
 followed.

 To ensure the integrity of the sample, some type of protective covering is required for filter
 handling and conditioning within the laboratory. A glass or plastic petri dish or petri slide,
 designed to hold these filters, is recommended. These are available from some filter manufac-
 turers and laboratory supply houses. The dish should be of suitable size (i.e., large enough to
 allow easy removal of the filter, yet small enough to prevent excessive movement within the
 container) and should have a tight-fitting lid to prevent intrusion of dust or loss of particles
 during handling at the analytical laboratory. A label should be affixed to the container to
 provide proper identification. A sufficient number of containers should be available to store
 filters during laboratory operations.

 4.3.6    Analytical Microbalance

 An analytical microbalance is required to weigh the sample filters. Its capacity should be ade-
 quate to weigh the sample filters (typically  100 to 500 mg). It should have sufficient room to
 weigh the type and size of filters used (i.e.,  46.2-mm diameter). The  microbalance must have a
 readability of no greater than ±1 ug and should have a repeatability of 1 ug. Readability is the
 smallest difference between two measured values that can be displayed by the  microbalance.
 Repeatability is a measure of the ability of a microbalance to display the same  result in
 repetitive measurements of the same weight under the same measurement conditions.

 The precision of mass measurements for unexposed filters based on replicate weighings will be
 greater than the microbalance's repeatability. The balance must be calibrated at installation and
 checked immediately before each weighing session. Electrostatic charge buildup may need to be
 neutralized in the microbalance's weighing chamber and on individual filters. Refer to Section
 9.6 and the balance manufacturer's instruction manual and technical  notes. Using microbalances
 with optional data input and output capabilities and automatic tare capabilities is recommended
 to reduce potential data entry errors and the time required to  weigh a given filter.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies                    January 2016   Page 9 of 10
 NOTE: Automated filter weighing and processing systems (e.g., PM2.5 autohandlers) are
 commercially available and can be purchased by monitoring organizations for PM2.5 laboratory
 operations. If procured, the monitoring organization must ensure that the autohandler's
 weighing operations and data recording capabilities meet all the PIVb.s requirements specified in
 40 CFR Part 50, Appendix L.
 4.3.7    Calibration Reference Standards

 Mass reference standards (or balance check weights) should be used to verify proper micro-
 balance operation as part of normal QC operations. These standards should be ASTM Class 0,
 Class 1, or Class 2, traceable to NIST. The individual weight tolerances for the ASTM mass
 reference standards are given in Table 4-2; balance check weights with an individual tolerance
 of no more than 0.025 mg should be utilized in the laboratory (ASTM 1993b)u .  See
 Section 9.7 of this document for more information regarding the mass reference standards and
the terminology associated with their certification.

 The mass reference standards should be selected so as to bracket the maximum and minimum
 expected filter weights by 100 mg  (e.g., 300 to 500 mg, given that the mass range of a typical
 46.2-mm filter is from 360 to 380 mg). Blank filter weights may occasionally change, so if
 purchasing filters from a new producer, check the blank filter weight to ensure the reference
 standards will bracket blank as well as loaded filters. Also, some microbalance designs only
 require calibration at their full scale weight; therefore, purchase of a mass reference standard at
 that weight should also be considered to verify full scale calibration. For the  same reasons,
 consult the microbalance manufacturer, and if possible, the manufacturer's microbalance
 operating instructions.

 Table 4-2. Tolerance specifications for Mass Reference Standards

            ASTM Class           Individual weight tolerance (mg)
                 0                           0.005
                 1                           0.010
                 2                           0.025
    NOTE: The microbalance's resolution and repeatability are smaller than the tolerance of the
    most accurate classes of mass reference standards. However, the reported uncertainty of a
    mass reference standard recalibrated at NIST-traceable or NVLAP-accredited laboratories
    will be smaller than the individual tolerance of the standard. The accuracy of the gravimetric
    analysis may be limited by the uncertainty of the standards rather than by the
    microbalance's characteristics.
11 llUfiL//www^^                                      reference.^}!
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 4.0 Procurement of Equipment and Supplies
            January 2016   Page 10 of 10
 4.3.8    Temperature and Relative Humidity Logger

 Temperature and relative humidity control in the weighing room are some of the most crucial
 criteria that must be met and recorded when weighing PM2.5 filters. A temperature and relative
 humidity logger is required in the laboratory to measure and record environmental conditions in
 the weighing room to provide evidence that weighing conditions have been met. The logger
 may be stand-alone or, preferably, networked into a data logging system. The logger should also
 be capable of continually logging data,  at a minimum, of every five minutes over at least a
 seven-day period. A log that records data every minute is encouraged. The logger should be
 recertified annually by the vendor.

 Minimum performance specifications are as follows:
      Relative humidity:

      •  measurement range of 20 - 50% RH
      •  readable to the nearest 0.5% RH
      •  accurate to within 2% RH
      •  five-minute logging
      •  NIST-traceable
Temperature'.

•  measurement range of 18 to 25°C
•  readable to the nearest 0.1°C
•  accurate to within 2°C
•  five-minute logging
•  NIST-traceable
     NOTE: Some loggers display rolling 5- or 15-minute averages for temperature and
     humidity. Actual raw data, and not rolling average data, needs to be used to determine
     room conditioning compliance.

 4.3.9    Audit Equipment
 Standards to audit laboratory equipment (the microbalance, and temperatures/humidity devices)
 should be procured. The audit standards must be NIST-traceable. The standards must not be the
 same equipment used for routine verification checks, although both sets of equipment may be of
 identical makes and models, and referenced to the same NIST primary standards.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.5.0 Installation                                              January 2016    Page 1 of 4


 5.0    Installation

 5.1    Siting Requirements

 As with any type of air monitoring study in which the sample data are used to draw conclusions
 about a geographic area, the validity of those conclusions depends on the representativeness of
 the sampling data. Therefore, an initial goal of a PIVfo.s monitoring project is to select a safe and
 secure site where the PIVfo.s measurements will be representative of the monitoring objectives
 for that site.

 5.1.1    Spatial and Temporal Considerations for Sites

 Spatial scale considerations are important in PM2.5 sampler siting. Spatial scales may range
 from small to large areas (i.e., micro, middle, neighborhood, urban, to regional scales). Whether
 the potential impact of particulate pollution is generated by a local or general source category
 affects the size of the spatial monitoring scale. In addition, the siting of the samplers within a
 monitoring network should reflect whether the expected impact will be limited to a small area (a
 few city blocks) or will  extend to larger areas  (metropolitan or rural).

 For temporal considerations, most interest focuses on either an annual geometric mean concen-
 tration or a 24-hour average concentration. Because siting of a PIVh.s sampler requires con-
 sidering the prevailing wind direction, a sampler sited for monitoring trends in air quality over a
 period of a year is not necessarily ideal for measuring 24-hour concentrations. Thus, the choice
 temporal considerations may influence sampler location. These  spatial and temporal aspects of
 network design and optimum site exposure are more completely explained in 40 CFR Part 58,
 Appendix D (EPA  1997a), and in the guidance document for network design and optimum site
 exposure for PM2.5 and PMio published by the U.S. EPA OAQPS (EPA 1997b).

 5.1.2    Sampler Location

 Although spatial and temporal scales should be considered in site selection, the following
 sampler location requirements (40 CFR Part 58, Appendix E) must be observed regardless of
 the scale:

    •   The PM2.5 sampler must have unobstructed air flow for a minimum of 2 meters in all
       horizontal directions and 1 meter vertical  spacing from supporting structures (i.e. a roof).
       Be sure to provide sufficient area for a collocated Federal Reference Method (FRM)
       sampler and for installation of a portable FRM Performance Evaluation Program (PEP)
       sampler.

    •   The sampler inlet must be placed at a height of 2 to 15 meters above ground level for
       neighborhood and larger spatial scales,  and at a height of 2 to 7 meters above ground
       level for middle scale and microscale sites.

    •   If a PM2.5 sampler is collocated with any other particulate matter sampler, the spacing
       between sampler inlets must be at least 2 meters apart for flow rates greater than 200
       liters/min or at least 1 meter apart for samplers having flow rates less than 200 liters/min

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.5.0 Installation                                              January 2016    Page 2 of 4

       to preclude airflow interference. In either case, the spacing between inlets must be no
       more than 4 meters. The heights of the inlets should be within 1 meter as measured in the
       vertical direction. A waiver allowing up to 10 meters horizontal distance and up to 3
       meters vertical distance (inlet to inlet) between a primary and collocated sampler may be
       approved by the Regional Administrator for sites at a neighborhood or larger scale of
       representation. This waiver may be approved during the annual network plan approval
       process as described in 40 CFRPart 58, Appendix A, Section 3.2.3.4.

 5.1.3    Safety

 Place the PM2.5 sampler used for routine sampling where the operator can reach it safely
 regardless of weather conditions.  If the sampler is located on a rooftop, care should be taken
 that the operator's personal safety is not jeopardized by a slippery roof surface during inclement
 weather. Consideration also should be given to the fact that routine operation (i.e., calibrations,
 sample filter installation and recovery, flow checks, and audits) involves transporting supplies
 and equipment to and from the monitoring site.

 5.1.4    Electrical Considerations

 In 40 CFR Part 50 Appendix L (EPA 1997c),  a PM2.5 sampler is required to operate at 105 to
 125 volts alternating current (AC) and a frequency of 59-61 hertz (Hz). The sampler may draw a
 higher current when the pump starts, possibly necessitating a slow-blow fuse. Although PM2.5
 samplers are required to indicate power interruptions, every effort should be made to provide a
 stable power source for the monitoring site. The site should be able to provide sufficient power
 for the primary sampler, a collocated sampler, and a PEP sampler.

 5.1.5    Security

 The security of the sampler itself depends largely on its location. Rooftop sites with locked
 access and ground-level sites with fences are common. Fences should be of chain-link or similar
 construction so as not to impede air flow across the site. The inlet of the sampler should extend
 above the top of the fence. In all cases, the security of the operating personnel, as well as the
 sampler, should be considered.

 5.2    Sampler Installation Procedures

 5.2.1    Receipt of Sampler

 Some sampler manufacturers may have time limits  for when they will accept damaged or
 inferior samplers back for repair.  Therefore, it may be important to inspect samplers within this
 warranty period to avoid additional repair costs.

   •   Establish an individual logbook for each sampler received.

   •   Upon receipt of a PIVb.s sampler from the manufacturer, visually inspect the sampler to
       ensure that all components are accounted for.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.5.0 Installation                                              January 2016    Page 3 of 4

   •   Compare the equipment delivered with the items listed on the enclosed packing slip.
       Notify the manufacturer immediately of any missing or damaged equipment.

   •   Read the manufacturer's instruction manual and become familiar with the sampler's
       operating and calibration procedures.

   •   Assemble the sampler in the monitoring organization's certification laboratory or
       maintenance facility ("shop") according to the manufacturer's instructions, except do not
       put oil in the impactor well (if using a WINS).

 5.2.2      Initial Evaluation of Samplers

 The following activities can most conveniently take place in an indoor, equipment certification
 laboratory environment. Field measurements personnel should conduct these tests.

   •   Power on the sampler. Check to be sure that the pump, mechanical components, clock,
       and electronic displays are functioning properly.

   •   Perform a leak check according to the manufacturer's instructions. Each sampler's
       procedure will be somewhat different from the others. If using a WINS, the impactor
       should not have oil in it. A flow check filter should be placed in the filter cassette holder
       for this test.

   •   Perform checks of the temperature and pressure sensors according to the manufacturer's
       instructions and Sections 7.4.5 and 7.4.6 of this document.

   •   Perform a flow verification check according to the manufacturer's instructions and
       Section 7.4.7 of this  document.

   •   For sequential samplers, check that the timing and sequencing functions work properly.

   •   If this is the first sampler model of its type received,  conduct data downloading exercises
       to become familiar with the process. If it is a sequential sampler, operate it for at least 3
       consecutive days to test the filter exchange system. Test the sampler in this manner for
       more than 3 consecutive days, if resources allow.

   •   If the separator is a WINS, and if oil has been added  to the impactor well during any of
       the acceptance testing checks, remove it before transit to the field  site.

 5.2.3     Setup at Sampling Site

   •   Carefully transport the sampler to the sampling site.

   •   Secure the sampler in its predetermined location. The legs of the sampler can be bolted to
       angle aluminum bars. The bars can be held in the soil with metal stakes or they can be
       nailed to wooden beams if the sampler is on the surface of a roof. Keep the sampler level.
       Allow space for the installation of a FRM PEP sampler and possibly for a collocated

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.5.0 Installation                                               January 2016     Page 4 of 4

       sampler12. As described in 40 CFR Part 58, Appendix A, collocated or PEP monitors
       must be within 4 meters of each other and at least 2 meters apart for flow rates greater
       than 200 liters/min or at least 1 meter apart for samplers having flow rates less than 200
       liters/min to preclude airflow interference. A waiver allowing up to 10 meters horizontal
       distance and up to 3 meters vertical distance (inlet to inlet) between a primary and
       collocated sampler may be approved by the Regional Administrator for sites at a
       neighborhood or larger scale of representation.

 5.2.4     Field Evaluation

 It is expected that prior to field implementation the sampler(s) has been evaluated and
 "checked" to ensure it is operating properly.  However, it may be appropriate to implement a
 few non-regulatory checks to  ensure any technical issues not discovered in initial evaluations
 are identified and corrected prior to regulatory sampling.

    •   Check all tubing for cleanliness, crimps, cracks, or breaks.

    •   Plug the power cord into a line voltage outlet.  The use of waterproof interlocking
       electrical connectors is  recommended to ensure operator safety and to avoid shorts and/or
       power interruptions. Do not allow any electrical connections to become submerged
       during periods of inclement weather.

    •   Ensure timer is programmed with the correct date, time, and sampling frequency. For
       SLAMS, clocks must be set to local standard time13.

    •   Perform a leak test with a filter in the filter cassette holder. Do not use the filter for
       subsequent sampling.

    •   Allow the system to run and equilibrate to ambient conditions. This may take about 15
       minutes.

    •   Perform checks of the temperature and pressure sensors.

    •   Perform a flow rate calibration/verification according to the manufacturer's instructions.

    •   Correct any problems before proceeding. Document all procedures and checks. The
       sampler is now ready for routine use.
12 Collocated QA samplers are required at 15% of sites for each primary method designation so not every site will
need to have additional space for this monitor.  However, every site will need to space for the PEP monitor.
13 40CFRPart58AppN

                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page 1 of 18


 6.0    Calibration Procedures


 6.1    Overview

 Calibration is defined as:

    comparison of a measurement standard or instrument with another standard or instrument
    to detect, correlate, report, or eliminate by adjustment any inaccuracy of the compared14.

Verification Versus Calibration

Since the term calibration is associated with an adjustment in either the instrument or software,
these adjustments should be minimized as much as possible.   Sometimes performing frequent
adjustments to provide the "most accurate data possible" can be self-defeating and be the cause
of additional measurement uncertainty.  For example, adjusting an instrument based upon a
standard that might be degrading or contaminated may actually cause data to be farther from the
true concentration.  Therefore, quality control procedures that include multi-point verifications
are  considered "checks without correction" and are used to ensure the instruments are within the
calibration tolerances. Usually these tolerances have been developed so that as long as the
instrument is within these tolerances, adjustments do not need to be made. However,
verifications should be implemented at reasonable frequencies to avoid invalidating significant
amounts of data. In this section it is assumed that multi-point verifications will take place and are
used to determine if a calibration (with adjustment) is necessary.

     NOTE: When the term "calibration" is used in  the remainder of this section, it is
     assumed that a multi-point verification is initially performed and the operator has
     concluded that calibration (adjustment) is necessary.

     NOTE: EPA does not recommend post-processing of data to "correct" for data
     failing one-point or multi-point verifications.  For example, if after failure of a flow
     rate verification and a subsequent verification and calibration found that data was
     biased by 15%, the previous routine data up until the last acceptable flow rate check
     is not adjusted by 15% and reported.  Based upon validation criteria, the data is
     either reported as initially measured or invalidated.

 Before a PIVb.s monitoring program is undertaken, all sampling and analysis equipment must be
 properly calibrated. Careful, accurate calibrations of sampling instrumentation and associated
 equipment are fundamental for any monitoring network. The suggestions and requirements
 specified in this section may serve as initial acceptance checks; further checks may be necessary
 if acceptance specifications are not met. All data and calculations should be recorded in a
 calibration logbook or on calibration data sheets. A separate logbook or a separate section of a
 logbook should be used for each apparatus and sampler in the program.  Once the initial
 calibration of a PM2.5 sampler's sensors is established, a schedule of field performance checks
 should be maintained to verify that the sensor calibration has not changed. Note that the
 calibration activity includes first determining if the sensor reading is still within the acceptance
14 Il!!j3://wwwjiij^^

                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page2of18

 limits of the calibration (multi-point verification). Whenever field performance checks of the
 sampler's flow rate, external or internal leakage rate, temperature, or pressure systems show
 out-of-tolerance conditions, the operator is cautioned to double-check the equipment, the
 connections, and the sampler before proceeding with recalibration. It is possible that the
 sampler is not out of calibration and considerable time and effort could be spent unnecessarily.

 According to  40 CFR Part 50, Appendix L, Section 10 (EPA 1997), the PM2.5 sampler flow rate
 measurement system must be calibrated in terms of the actual volumetric flow rate (Qa) at
 prevailing ambient conditions of temperature and pressure, as opposed to the standard volu-
 metric flow rate (Qstd), which is measured or corrected to U.S. EPA-standard conditions of
 temperature and pressure. The flow rate measurement system consists of a dry gas meter or
 other flow sensor, a temperature sensor, a pressure sensor, and associated devices which, as
 needed,  make control adjustments to maintain the set point of the volumetric flow rate in
 response to variations in temperature and pressure. Refer to the operating manual for the PIVb.s
 sampler for specific instructions on how to enter calibration data using the keypad and how to
 make other adjustments. Measured PM2.5 concentrations are also reported in terms of the actual
 volume  at ambient conditions of temperature and pressure.  Therefore, no conversions between
 Qa and Qstd are generally necessary. However, if the calibration for your flow rate standard or
 transfer  standard is expressed in terms of Qstd, conversion to Qa is required. The calculations
 necessary to convert these flow rates are presented in Section 6.2, Equations (6-1) and (6-2).

 This section presents the following aspects of calibration procedures:

   •   A discussion of Qa and Qstd flow rate designations and their applicability in a PIVfo.s
       monitoring program (Section 6.2).

   •   Basic calibration procedures and associated calculations for calibration of the sampler's
       flow rate measurement system (Section 6.3).

   •   Discussion of flow rate transfer standards  and calibration  equipment (Sections 6.3.2 and
       6.3.3).

   •   Basic calibration procedures for the sampler's temperature and pressure measurement
       systems (Sections 6.4 and 6.5).

   •   Procedures for conducting  external and internal leak checks (Section 6.6 and 7.4).

   •   Sampler calibration frequency requirements (Section 6.7).

 More detailed information on calibration  and operation procedures can be found in each sampler
 manufacturer's operating instruction manual. These instruction manuals are a required part of
 the manufacturer's application for designation as an EPA Reference or Equivalent Method and
 as such have been reviewed and accepted for their designated sampler.
                                               2.12-PM2.5 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page3of18

 6.2   General Aspects of Flow Rate Measurement and PIVh.s Sampler Calibration

 As discussed in Section 3.0, "Summary," a reference or Class I equivalent PM2.5 sampler
 includes a specially designed sample air inlet, a PIVfo.s separator (e.g., WINS or VSCC™), and a
 sample flow rate control system. The particle-size discrimination characteristics of both the inlet
 and the separator are critically dependent on specific internal air velocities; a change in velocity
 will result in a change in the nominal particle size collected. These velocities are determined by
 the actual volumetric flow rate (not the mass flow rate) through each device. Therefore, the
 actual volumetric flow rate through the sampler's inlet and separator must be maintained at a
 constant value that is as close as possible to the design flow rate specified for the  sampler. The
 design flow rate for a given sampler is specified in the sampler's operation or instruction
 manual. Upper and lower tolerance limits for the sampler's flow rate are specified in 40 CFR
 Part 50, Appendix L, and are + 5% of the design flow rate.

 As mentioned previously, the calibration of the sampler's flow rate measurement system must
 be performed in terms of actual volumetric units (Qa or  Va). Results must be recorded in the
 same units. However, previous measurements for PMio  were based on standard units (Qstd or
 Vstd)', thus, the  certification for some calibration standards may be in terms of standard
 volumetric flow rate units (Qstd). Therefore, it is important to understand the definition for each
 of these two types of volumetric units and the distinction between them. Before calibration
 procedures are initiated, the monitoring organization personnel should review the following
 flow rate definitions:

    •  Qa—Actual volumetric air flow rates that are measured and expressed at existing
       conditions of temperature and pressure are denoted by Qa (Qactuai). Typical units are
       L/min and m3/min. Inlet design flow rates for PM2.5 samplers are always given in actual
       volumetric flow units.

    •  Qstd—Air flow rates that have been adjusted to EPA-standard conditions of temperature
       and pressure  (25 °C or 298 K and 760 mmHg or 101 kPa) are denoted by Qstd (Qstandard).
       Typical units are L/min and mVmin. Standard volume flow rates are often used by
       engineers and scientists because they are equivalent to mass flow rate units. Prior to
       1997, standard volumes (derived from standard volume flow rates and the total time of
       sampling) were also required in the calculation of mass concentration (ug/m3) in
       reporting PMio measurements.

 These Qa and Qstd Row rate units must not be confused or interchanged. If necessary, the flow
 rate units can be converted, provided the temperature and pressure are known. The following
 conversion formulas can also be used for average flow rates ( Qa  and  Qstd ) over a sampling
 period by substituting average temperature (  Ta  ) and pressure  (  Pa ) over the sampling
 period.
                                 Qstd = Qa(Po/Pstd)(Tstd/Ta)                             (6-1)

                                 Qa = Qstd(Pstd/Pa)(Ta/Tstd)                             (6-2)

 where
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page4of18

        Qstd  = standard volume flow rate, standard m3/min
         Qa  = actual volume flow rate, actual m3/min
         Pa  = ambient barometric pressure, mmHg (or kPa)
        Pstd  = EPA standard barometric pressure, 760 mmHg (or 101 kPa)
        Tad  = EPA standard temperature, 298 K (25 °C + 273)
         Ta  = ambient temperature, K (ambient °C + 273).

 Any gas flow rate measured or expressed in actual volumetric units (Qa) is always associated
 with a particular temperature and pressure of the gas. If either the temperature or the pressure
 changes, the volumetric flow rate will also change, even though the mass flow rate of the gas
 remains constant. Therefore, when the flow rate is measured at different locations in the
 sampler, the volumetric flow rate observed will be different if either the temperature or the
 pressure is different at the various locations. For example, when a flow calibration device is
 connected to the sampler inlet, the pressure of the air flow measured by the flow calibration
 standard will be the ambient barometric pressure. However, the pressure of the flowing air
 measured by the sampler's flow measurement system will be somewhat lower than the ambient
 barometric pressure because all flow calibration devices cause some pressure drop. If this
 pressure drop is negligible (<1  percent or <4 in of H2O), then there is no problem; the pressure
 can be considered the same for both measurement systems. But if the pressure drop is
 significant (>1 percent or >4 in of H2O), then the volumetric flow rate measured by the two
 systems will be different, and this difference should be taken into account when comparing
 them  Sampler flow rate measurement systems may or may not automatically correct for
 this pressure change. If not, one of the flow rate measurements must be corrected to the same
 pressure as the other flow rate before they can be compared. This correction is accomplished
 using the following formula:
                                 Qi = Q2 (P2/Pi)(Ti /T2)                             (6-3)
 where

         Qi  = actual volume flow rate at the pressure and temperature at the first measurement
               point, mVmin
         Q2  = actual volume flow rate at the second measurement point, m3/min
         Pi  = pressure at the first measurement point, mmHg (or kPa)
         P2  = pressure at the second measurement point, mmHg (or kPa)
         Ti  = temperature at the first measurement point, K (°C + 273)
         T2  = temperature at the second measurement point, K (°C + 273).

 Note:  In many cases the temperatures will be  identical, so that Ti=T2 and the (Ti/T2) term will
 equal  1.

 6.3    Calibration of the Sampler Flow  Rate Measurement System

 A full, detailed, EPA-approved calibration procedure, tailored specifically to each commercially
 available PIVfo.s sampler, is contained in the operating or instruction manual associated with
 each sampler designated as a reference or equivalent method under 40 CFR Part 53. Follow that
 specific procedure carefully and thoroughly to calibrate the sampler.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures
January 2016     Page 5 of 18
                                       CAUTION

           Do not calibrate the flow rate measurement system without first
           performing a leak check and ensuring that the temperature and pressure
           measurement systems are in calibration. If there is doubt, calibrate the
           temperature and pressure sensors (refer to Sections 6.4 and 6.5) before
           calibrating the flow rate measurement system.
 This section provides additional precautions, guidance, and information on flow rate standards
 and calibration that may not be contained in the specific instruction manual procedure. It also
 presents a generic outline of the general procedure used to calibrate the flow rate measurement
 systems of commercially available samplers. This information can be used to augment the
 sampler-specific procedures.

 6.3.1    General Requirements and Guidance

Checking to verify that a sampler's flow rate measurement system still reads within the
acceptance limit of the previous calibration should be done:

   •   according to the sampler manufacturer's operating instructions;
   •   at the field site;
   •   every 30 days (minimum); or
   •   following major electrical or mechanical maintenance, such as replacement of a circuit
       board or rebuilding of the sampling pump assembly.

 If the verification fails, then recalibration and/or maintenance is required.  Before beginning a
 flow rate calibration, the operator should repeat the flow rate verification to be certain there
 were no leaks in the connections between the flow rate verification device and the sampler inlet
 and that the flow rate data were reduced and interpreted correctly. In addition, the operator
 should ensure that the collector's ambient temperature and pressure measurement systems are
 responding accurately; if not, these systems should be recalibrated (or replaced and calibrated)
 before performing a flow rate calibration.
     Multi-point calibrations and single-point verifications of each sampler's flow rate must be
     performed periodically as defined in the PM2.5 Validation Template15to establish
     traceability of subsequent flow rate measurements to an authoritative flow rate standard.
     Calibration with a flow rate standard (or  a transfer standard) that is certified against a
     National Institute of Standards and Technology (NIST)-traceable standard (see Section
     6.3.3 regarding NIST traceability) transfers the NIST traceability to the sampler's flow rate
     measurement system. This NIST traceability is required by Appendix A of 40 CFR Part 58
     Sec. 2.6.3.
15 http://www3.cpa.gov/ttJT/aintJc/qapolliitantJitml
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016     Page6of18
 2.   PM2.5 samplers may employ various types of flow rate measurement devices. The specific
     calibration standard and procedure used for calibration or verification of the sampler's flow
     rate measurement device will vary depending on the type of flow rate measurement system
     employed (40 CFR Part 50, Appendix L, Sec. 9.2.1). The sampler's operation or instruction
     manual identifies one or more types of flow rate or volume standards recommended for
     calibration and provides a detailed calibration procedure.

 3.   Calibration of the sampler's flow rate measurement system must be in units of the actual
     ambient volumetric flow rate (Qa) (40 CFR Part 50, Appendix L,  Sec. 9.2.1).

 4.   The sampler flow rate measurement system must be calibrated or verified by installing an
     unused filter in the filter holder, removing the sampler inlet, and connecting the flow
     adapter and flow rate standard to the sampler, in accordance with the instruction manual, so
     that the flow rate standard accurately measures the sampler's flow rate. The operator shall
     verify that no leaks exist between the flow rate standard  and the sampler (40 CFR Part 50,
     Appendix L, Sec. 9.2.3).

 5.   The calibration relationship between the flow rate (in actual L/min) indicated by the flow
     rate standard and by the sampler's flow  rate measurement device  must be established or
     verified in accordance with the sampler's operation or instruction manual. Temperature and
     pressure corrections may be required for some types of flow rate standards (40 CFR Part
     50, Appendix L, Sec. 9.2.4). Consult the operations manual for the particular flow rate
     standard for guidance on the nature and  calculation of any corrections that may be required
     for the standard.

 6.   Monthly verification of the sampler's flow rate  shall consist of one flow rate measurement
     at the sampler's operational flow rate (40 CFR Part 50, Appendix L, Section 9.2.4). A full
     three-point calibration must be carried out whenever a one-point verification indicates that
     the sampler's flow rate  measurement system differs by ±4 percent or more from the flow
     rate measured by the flow rate standard  and greater than  5% of the design flow rate (40
     CFR Part  50, Appendix L,  Sec. 9.2.5). The one-point verification must be repeated after the
     three-point calibration to double-check that the  sampler operates properly at the design
     flow rate of 16.67 L/min following  the calibration.

 7.   Calibration of the sampler's flow rate measurement device must consist of at least three
     separate flow rate measurements (a multi-point  calibration), evenly  spaced within the range
     of-10 to +10 percent of the sampler's operational flow rate (40 CFR Part 50, Appendix L,
     Section 9.2.4). The sampler's flow control system shall allow for operator adjustment of
     the operational flow rate of the sampler  over a range of at least ±15  percent of the targeted
     flow rate (40 CFR Part  50, Appendix L, Section 7.4.2). Calibration procedures for the
     various samplers call for the results of three different flow rates to be keyed in to establish
     the calibration. The sampler's instruction manual will provide additional guidance on flow
     rate adjustment.

 8.   Following a calibration, the flow adapter and flow rate standard are disconnected from the
     sampler, and the sampler's inlet is carefully reinstalled. The flow check  filter remains in
     place. The sampler's flow rate must then be checked under normal operating conditions,

                                              2.12-PM25  Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page7of18

     using the sampler's flow rate measurement system. The operator should generate flow and
     view the displayed flow rate on the sampler's screen.  Ensure the displayed flow rate
     differs by no more than ±2 percent from the required operational value of 16.67 L/min (40
     CFR Part 50, Appendix L, Sec. 9.2.6). If the sampler's flow rate meets this criterion, the
     filter is then removed and the sampler is returned to service. However, if the sampler's
     displayed flow rate is more than ±2 percent different, the sampler's flow rate must be
     adjusted again to the specified flow rate.

 6.3.2     Flow Rate Calibration Standards

 Flow rate standards used for calibrating or verifying the sampler's flow rate measurement
 device must have an accuracy of ±2 percent. This flow rate standard must be a separate, stand-
 alone  device, and may be designed to connect to the flow adapter device. The flow rate standard
 must have its own certification and must be traceable to a NIST primary standard for volume or
 flow rate (40 CFR Part 50, Appendix L, Section 9.1.2). See Section 6.3.3 of this document for
 more information on NIST traceability. Consult the technical literature and the vendor's
 descriptions of various flow rate standards and their performance under various weather
 conditions before making a decision to purchase one.

 All flow rate calibration standards should be fully equilibrated to the temperature of the air of
 the instrument whose flow rate is to be measured. This temperature equilibration can take up to
 an hour, depending on the temperature at which the standard was stored prior to moving it to the
 point of use. During this thermal equilibration period, the standard should be exposed to the
 prevailing air temperature, but it also should be protected from precipitation, wind, dust, solar
 heating, and other conditions that could affect its accuracy.

 Various types of flow rate standards that might be considered for sampler calibration follow.

    •   Bubble flowmeters—Although completely manually operated and timed bubble
        flowmeters are available, a semiautomatic instrument such as the Gilian* flowmeter is
        easier to use and is more likely to provide uniform flow rate measurements. Because the
        liquid used to form the bubbles is usually water-based, a water-vapor correction may be
        required to compensate for a volume increase due to evaporation of the water in the
        instrument. Furthermore, the instrument cannot be used below 0 °C because the soap
        solution will freeze. Bubble meters are primary standards, but are not conducive for use
        in the field.

    •   Piston flowmeters—Automatic dry-piston flowmeters such as the BIOS* flowmeter are
        relatively easy  to use  and not subject to the water-vapor correction or the liquid freezing
        limitation of the bubble flowmeters. However, they may still be subject to a lower
        operational temperature limit.

    •   Mass flowmeters—Although mass flowmeters may be considered as calibration
        Mention of commercial products should not be interpreted as endorsement. This product is given as a
typical and perhaps well-known example of the general class of instruments; other instruments in the class are
available and may be fully acceptable.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                  January2016    Page8of18

        standards, they have a number of shortcomings in this application. They have a high
        pressure drop, which could change the sampler flow rate or otherwise affect the
        operation of the sampler. Mass flowmeters also require conversion of the measured mass
        flow rate to the actual volumetric flow rate for comparison to the PIVb.s sampler's
        displayed value. The flow rate measured by the sampler may also have to be corrected to
        account for the significant pressure drop introduced by the mass flowmeter itself.
        Finally, mass flowmeters are subject to zero or span drift, particularly over the wide
        range of temperatures encountered at field sites, so their accuracy should be reverified
        frequently over a wide temperature range.

    •   Orifice devices—Orifice devices are simple and reliable, but they too require
        corrections  for temperature and pressure and may have a significant pressure drop. They
        also require a very accurate and sensitive differential pressure measurement device such
        as a manometer or an aneroid differential pressure gauge, whose accuracy must be
        reverified frequently. Readings from electronic micromanometers should be cross-
        checked against a water manometer at  some frequency in order to verify the standard is
        operating within acceptable limits.

    •   Laminar flow elements—Laminar flow elements generally have a lower pressure drop
        than orifice devices, but otherwise have the same requirements and disadvantages.

    •   Wet test meters—Wet test meters are generally not practical for field use, but may be
        used for laboratory calibrations.

    •   Dry gas meters—Dry gas meters may be considered for field calibration.  Some meters
        may have a substantial pressure drop. They should be checked for leaks and mechanical
        problems and be recertified at least annually.

 6.3.3   NIST Traceability and Certification of Flow Rate Standards

 The flow rate standard used for flow rate calibration should have its own certification and
 should be traceable to other standards for volume or flow rate, standards that are themselves
 NIST-traceable. It is recommended that the "other standards" be either those of a commercial
 calibration laboratory or those primary standards maintained by the SLT agency or EPA
 Region. The greater the number of calibration steps needed to link a measurement to a NIST
 standard, the greater the degradation of quality of this type of traceability. Thus, the "other
 standard" should be at least as accurate (EPA suggests more accurate) and stable as the flow
 rate standards to be made traceable and should be no more than two traceability steps removed
 from an actual  NIST standard. A calibration relationship for the flow rate standard, such as an
 equation, curve, or family of curves, should be established that is accurate to within 2 percent
 over the expected range of ambient temperatures and pressures at which the flow rate standard
 is expected to be used. The flow rate standard  must be recalibrated or reverified and recertified
 at least annually16.  The EPA's Quality Assurance Handbook for Air Pollution Measurement
 Systems, Volume II should be consulted for further information on the calibration of primary
 and secondary  standards for flow measurements and the hierarchy of standards.
16 40 CFR Part 50 Appendix L Se. 9.2.2

                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                   January2016    Page9of18

The actual frequency with which this recertification process must be completed depends on the
type of flow rate standard—some are likely to be much more stable than others. The best way to
determine recertification requirements is to maintain a control chart (a running plot of the differ-
ence or percent difference between the flow rate standard and the NIST-traceable primary flow
rate or volume standard) for all comparisons. In addition to providing excellent documentation of
the certification of the standard, a control chart also gives a good indication of the stability of the
standard. If the two standard deviation control limits are close together, the chart indicates that
the standard is very stable and could be recertified less frequently. The minimum recertification
frequency is every 365 days and once a calendar year. On the other hand, if the limits are wide,
the chart would indicate a less stable  standard that should be recertified more often. A particu-
larly unstable standard may be unusable no matter how often it is recertified.

It is also recommended that all transfer standards being sent out for certifications (or being
checked/certified in-house) be checked against a specific piece of stationary laboratory
equipment in house to observe the "As sent" and the "As returned" conditions of the standard.
Monitoring organizations have experienced what they have considered inferior certifications
from vendors,  so it is a good idea to verify standards that are returned prior to putting them into
service.

 6.3.4   Outline of Generic Flow Rate Calibration Procedure

 A fully detailed, EPA-approved flow rate calibration procedure, tailored specifically to each
 commercially available PIVfo.s sampler, is contained in the operations manual associated with
 each sampler that is designated as a reference or equivalent method under 40 CFR Part 53.
 Follow that sampler-specific procedure carefully and thoroughly, in conjunction with the
 guidance in this section, to calibrate the sampler. The following steps outline a general
 calibration procedure. Figure 6.1 can be used as an example worksheet to record this
 information

     NOTE: Upon initial installation at the field site, ensure that the sampler is operational and
     fully warmed up before proceeding. Depending on weather conditions, the sampler should
     run for a minimum of 10 to 15 minutes before starting the flow rate calibration process.

 1.  Before performing a flow rate calibration, make sure that an external leak check has been
     performed and passed, and that the sampler's temperature and pressure sensors have been
     checked and have met acceptance criteria. If any of these checks fail, perform corrective
     actions prior to starting the flow rate calibration procedure below.

 2.  Equilibrate the selected flow rate calibration  device to ambient temperature conditions
     according to guidance provided in Section 6.3.2.

 3.  Install  a filter cassette with an unused 46.2-mm filter in the sampler. This filter should meet
     all specifications for PM2.5 sampling, but it does not need to be pre- or post-weighed. This
     filter must not be used for sampling but can be used several more times for calibration
     purposes, provided particulate loading on the filter is minimal.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                 January2016    Page10of18

 4.   Remove the inlet from the sampler. Place the flow calibration device on the sampler down-
     tube using a flow adapter device, if necessary. Ensure that any valves in the flow adapter
     are open so that flow through the sampler is unrestricted.

 5.   Place the sampler in calibration mode according to the instructions in the manufacturer's
     operating manual.

 6.   Follow the instructions in the manufacturer's operating manual for performing the multi-
     point flow calibration.

 7.   Once calibration is complete, turn off the sampler pump, remove the filter/cassette, remove
     the flow calibration device (and flow adapter device, if applicable), and replace the sampler
     inlet.

 8.  The sampler flow rate is now calibrated. Refer to Section 7.4.7 for the procedure for flow
    rate verification, which will be used until the next flow rate calibration.

 6.4    Calibration of Sampler Temperature Sensors

 A fully detailed, EPA-approved calibration procedure for temperature sensors, tailored specifi-
 cally to each commercially available PM2.5 sampler, is contained in the operating manual
 associated with each sampler that is designated as a reference or Class I equivalent method
 under 40 CFR Part 53. Follow these specific procedure(s) carefully and thoroughly to calibrate
 the sampler's temperature sensors. It is recommended that a three-point temperature calibration
 of each temperature sensor be performed each year and that a one-point check of the ambient air
 temperature sensor be made monthly.

 This  section provides additional precautions, guidance, and information on temperature
 standards and calibration that may not be contained in the specific instruction manual
 procedure. The section also presents a generic outline of the general procedure used for
 calibration of the temperature measurement systems of commercially available samplers. This
 information should be used to augment the specific instruction manual procedure.

 6.4.1    General Requirements and Guidance

 1.   Multi-point (at least three temperature points) calibration followed by single-point
     verification of each sampler's temperature sensors should be performed annually to
     establish traceability of subsequent temperature measurements to an authoritative
     temperature standard. Calibration with a temperature standard that is traceable to a
     NIST-traceable primary  standard (see Section 6.3.3 regarding NIST-traceability) transfers
     the traceability to the sampler's temperature sensors. Because temperature affects the flow
     rate, NIST traceability of the sampler's temperature measurements is also required for flow
     rate measurement traceability.

 2.   PM2.5 sampler manufacturers use various types of temperature sensors. The specific
     calibration standard and  procedure used for calibration or verification of the sampler's
     temperature sensor may vary depending on the type of sensors used.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                January2016    Page 11 of 18

 3.   Multi-point calibration of the sampler's temperature measurement system must consist of at
     least three separate comparative temperature measurements, evenly spaced over the
     sampler's expected operational ambient temperature range. Both the ambient air and filter
     temperatures are monitored in order to assess filter temperature deviation from ambient
     temperature  and to set a flag should the filter temperature rise more than 5 °C above
     ambient temperatures and stay there for 30 consecutive minutes.

 4.   Section 3.4 of the EPA Quality Assurance Handbook for Air Pollution Measurement
     Systems,  Volume IV: Meteorological Measurements (EPA 2008) contains detailed guidance
     on calibration of temperature sensors and measurement systems (including a reference to
     the manufacturer's manual for adjustment of signal conditioning circuits), required
     calibration equipment, methods for calibrating the temperature, a discussion of AT
     calibrations,  and an example of a calibration report.

 5.   Ideally, temperature calibration should occur in the field to assess temperature sensor
     responses as they occur in normal operation. However, climate conditions may make this
     difficult and removal of the sampler to an indoor location may be preferable. Usually, the
     sensor element is removed from the sampler and its connecting wires are left attached to
     the sampler.  The space formerly occupied by the sensor should be plugged with a fitting to
     prevent any contamination of the sampling system and entry of ambient air.

 6.   Monthly verification of a sampler's temperature sensors responses should consist of one
     temperature  measurement made at the sampler's current temperature reading. This one-
     point verification may be substituted for a three-point  calibration, provided that a full
     three-point calibration is performed upon initial installation of the sampler and at least once
     per year thereafter, and the temperature measurement  system has met the ±2 °C accuracy
     requirement  in the previous three-point calibration. A  full three-point calibration should be
     performed whenever a 1-point verification indicates that the sampler's temperature
     measurement system differs by ±2 °C or more from the temperature measured by the
     temperature  standard. A  1-point verification of the ambient temperature sensor should be
     done following the three-point calibration to ensure the sampler has returned to its normal
     operating conditions.

 6.4.2    Temperature Calibration Standards

 The operations manual associated with the sampler should  identify one or more types of
 temperature standards recommended for calibration and provide a detailed calibration procedure
 for each type that is specifically designed for the particular sampler.

 The EPA Quality Assurance Handbook for Air Pollution Measurement Systems, Volume IV:
 Meteorological Measurements (EPA 2008), Section 3.4, provides information on calibration
 equipment and methods for assessing response characteristics of temperature sensors. The
 ambient air and filter temperature  sensors of a reference or Class I equivalent PM2.5 sampler
 must have a resolution of 0.1  °C and an accuracy of ±2 °C  over the range of-30 to 45 °C. The
 Handbook describes how to prepare three stable  thermal mass assemblies whose temperatures
 can be determined to about 0.1 °C. The thermal mass assemblies can be as simple as insulated
 vacuum bottles (i.e., thermos  bottles) containing pure water or ice, or they may be solid


                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                 January2016    Page12of18

 cylinders of aluminum metal. A good American Society for Testing and Materials (ASTM)- or
 NIST-traceable alcohol thermometer or thermocouple is also needed. The thermometer and the
 PM2.5 sampler sensor are both immersed in the thermal mass and allowed to equilibrate; the
 temperature readings are then compared.

 6.4.3    NIST-Traceability and Certification of Temperature Standards

 The temperature standard used for temperature calibration must have its own certification
 showing traceability to a NIST primary standard. A calibration relationship to the temperature
 standard (an equation or a curve) is established that is accurate to within 0.5 °C over the
 expected range of ambient temperatures at which the temperature standard will be used. The
 temperature standard should be reverified and recertified at least annually.

 The actual frequency of recertification depends on the type of temperature standard; some are
 much more stable than others. The best way to determine recertification requirements is to
 maintain a control chart.

 Additional reference sources concerning temperature measurements and calibration are: Liquid-
 in-Glass Thermometry (NIST 1976); Thermometer Calibration: A Model for State Calibration
 Laboratories (NIST 1986); NIST Measurement Services: Liquid-in-Glass Thermometer
 Calibration Service (NIST 1988); and The Calibration of Thermocouples and Thermocouple
 Materials (NIST 1989).

 6.4.4    Outline of Generic Temperature Calibration Procedure

 Both the ambient air and filter temperature sensors should be calibrated once per year. The
 ambient air sensor is located inside the shielded fixture on the outside of the PM2.5 sampler and
 is easy to unfasten and remove for comparison to a temperature transfer standard. It is possible
 to conduct the three-point calibration of the ambient sensor at the field site, although it may
 prove easier to remove the sampler to the field office to avoid weather problems and for
 convenience in preparing the temperature standards. The filter temperature sensor of reference
 or Class I equivalent PIVb.s samplers is located in the open space just below the filter cassette. It
 is threaded through the wall of the filter assembly section of the sampler, and removal of plastic
 or metal fittings is required to remove the sensor and its associated wiring. It is recommended
 that this sensor be calibrated in the field office. The temperature sensor housing, the sampler
 inlet, and the interior of the downtube can also be cleaned in the laboratory. Be careful when
 removing the filter temperature sensor—do not gall the fittings because this could start  an
 internal leak after installation. It is suggested that a sampler leak check be performed after
 reinstallation of the filter temperature sensor.

 Several steps to follow in calibrating ambient air temperature sensors are given below. Make
 frequent reference to the  operator's instruction manual for sampler-specific procedures  and
 instructions.

 1.   Remove the ambient temperature sensor from the aspirated radiation shield so that it can be
     placed in a constant  temperature bath while it  is still connected to the sampler's signal
     conditioner.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                January2016    Page13of18

 2.   Prepare a convenient container (such as an insulated vacuum bottle) for the ambient tem-
     perature water bath and the ice slurry bath. See Step 3 below. If complete immersion of the
     sensor is necessary, wrap it in plastic film so liquid can reach the point where the con-
     necting  wire(s) and the sensor interface without wetting them. Use partial immersion when
     possible, thus keeping the interface dry. If immersion should be avoided altogether, it will
     be necessary to use thermal masses of metal rather than those based on water. Refer to
     Section 3.4 of Volume IV of the EPA QA Handbook (EPA 2008). To further insulate the
     vacuum bottle, it can be positioned inside a larger 2-gallon insulated container that has
     been modified to allow wires or cables to enter the top.

     Keep the temperature changes relatively small and make comparative measurements in this
     order:   AMBIENT^COLD^AMBIENT^HOT^ AMBIENT.  The range of
     temperatures need only be as broad as that expected to contain all the ambient temperatures
     that will be experienced during the upcoming time period, generally a year. The range to be
     expected is site-specific. HOT may have to be 120 °F for Phoenix, AZ, but around 80 °F
     for a location such as Fairbanks, AK.

 3.   For the ambient bath, use an insulated bottle that was filled with tap or deionized water
     several hours earlier and allowed to equilibrate to ambient temperature. For the ice slurry,
     the ice should be made with distilled water and then crushed into pea-sized pieces and
     mixed with distilled water until an easily penetrable slurry state is reached. As long as ice is
     present in the slurry and the open end of the bottle is guarded from ambient air temperature
     fluctuations, the ice slurry temperature will be 0.0 ± 0.1°C.

 4.   Wrap the sensor(s) and a thermometer together with a rubber band. The thermometer bulb
     and the temperature sensor active site should be close together. Immerse the sensor and the
     attached thermometer in the ambient temperature bath. Use a cork or some other device to
     cover the open end of the insulated bottle and thus keep ambient air from circulating over
     the top surface of the water (or ice slurry mass). The bath liquid should be stirred to ensure
     the temperature is uniform; never use the thermometer or temperature sensor to stir the
     bath. Gentle stirring should continue, if possible, during the measurement process;
     however, do not stir while readings are being taken because this may introduce noise into
     the readings. Wait for the ambient thermal mass and the sensor/thermometer temperatures
     to equilibrate. Be sure successive temperature readings are stable (indicating equilibration
     with the ice slurry) before taking comparative readings.

 5.   For each thermal test condition or mass, in the order indicated in Step 2 above, make a
     series of five measurements, taken about a minute apart. Accurately read the meniscus of
     the thermometer. Use magnification if necessary to see the meniscus; avoid parallax errors.
     If the measurements made support the assumption of equilibrium, then average the five
     readings and record the result as the sensor temperature relative to the thermometer for hot,
     ambient, and cold, 0.0 °C relative to the ice slurry. Record all readings in the sampler
     notebook and/or on the monitoring organization's calibration form/spreadsheet.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                 January2016    Page14of18

 6.5   Calibration of Sampler Barometric Pressure Sensor

 Each reference or Class I equivalent PM2.5 sampler has a built-in atmospheric pressure sensor
 whose output is processed to allow control of the actual sampling flow rate to the design value
 of 16.67 L/min. This section provides information concerning the use of barometers and a
 general procedure for sampler calibration. Several types of barometers are available to measure
 barometric pressure, but the most common and user-friendly type of barometer is the digital
 pressure indicator. All barometers, digital and mechanical, must be recertified or calibrated to a
 NIST-traceable standard on a regular basis (e.g., yearly). The manufacturer's operating or
 instruction manual must be consulted for sampler-specific information on how to make
 adjustments to calibrate the pressure sensor.

 Due to the complications in setting up a 3-point barometric pressure calibration, EPA will
 accept a one-point verification with an independent NIST-traceable audit device (other than the
 device used for monthly routine verifications) to satisfy an annual calibration. The procedure
 suggested in Section 6.5.2 can be used for a multi-point calibration.

 6.5.1     General Requirements

 1.   As required in 40 CFR Part 50, Appendix L, the sampler must have the capability to
     measure the barometric pressure of the air surrounding the sampler over a range of 600 to
     800 mmHg. This measurement shall have a resolution of 5 mmHg and a NIST-traceable
     accuracy of ±10 mmHg.

 2.   According to ASTM standard D 3631 (ASTM 1977), a barometer can be calibrated by
     comparing it with a secondary standard traceable to a NIST primary standard.

 3.   Protect all barometers from violent mechanical shock and explosively sudden changes in
     pressure. A barometer subjected to either of these events must be recalibrated. Minimize
     the vertical and horizontal temperature gradients across the instruments. Locate the
     instrument so as to avoid direct sunlight, drafts, and vibrations.

 4.   Barometric pressure calibration is suggested at installation and on a one-point verification
     failure.

6.5.2    Outline of Generic Barometric Pressure Calibration Procedure

The barometric pressure sensor should be  calibrated once per year. The barometric pressure
sensor is typically located inside the PIVb.s sampler body and can be difficult to access. It is
possible to conduct the three-point calibration of the barometric pressure sensor at the field site,
although it may prove easier to remove the sampler to the field office to avoid weather problems
if a partial disassembly is required. As stated above, a 1-point pressure verification is acceptable.

Several steps  to follow in calibrating barometric pressure sensors are given below. This is a
general procedure; therefore, refer to the operator's instruction manual for sampler-specific
procedures and instructions. Descriptions of different types of barometers follow this section to
aid the operator in selecting and using barometric pressure standards for calibration and/or


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures                                 January2016     Page15of18

verification procedures.

 1.  Place the calibration standard aside, away from direct sunlight and extremes of temperature,
    and allow the unit to equilibrate to ambient conditions. Typical equilibration times for
    calibration standards are from 15 to 30 minutes depending on starting temperatures.

 2.  As the calibration standard equilibrates, find the small port on the sampler which leads to
    the barometric pressure sensor. The operator's manual can help locate and access this port.
    This port will be located inside or under the sampler where it is protected from the weather.

 3.  Setup the  sampler to show the calibration screen for barometric pressure calibration. The
    screen should contain the current barometric pressure measurement and an input mechanism
    to adjust the measurement. This screen will be used to adjust the  sampler measurement to
    the calibrator measurement during the calibration process. On some samplers, the operator
    may be required to manually adjust a potentiometer to adjust the  measurement.

 4.  Attach three short lengths of tubing to a "T" adapter and connect the first length of tubing to
    the pressure port.  The second length should be connected to the pressure fitting of a NIST-
    traceable barometer or pressure indicator, and the third length of tubing should be connected
    to a gas-tight syringe.

 5.  Draw back on the plunger to apply  a light suction until the NIST-traceable barometer reads
    about 100 mmHg below ambient pressure (e.g., 660 Hg if ambient pressure is 760 mmHg).
    Clamp off the hose with a tubing clamp or hemostats to prevent leakage and thus hold the
    pressure steady.

 6.  If the two numbers do not match within 10 mmHg or better,  adjust the sampler's barometric
    pressure display or potentiometer. Record the actual (calibrator measurement) and indicated
    (sampler measurement) values on a calibration form or field  logbook.

 7.  Repeat this process at ambient conditions and at a point 30 to 100 mmHg higher than
    ambient pressure  recording all values throughout the process.

 8.  If the sampler cannot be brought into satisfactory agreement with the NIST-traceable
    barometer, troubleshooting and repairs may be required or the pressure sensor may have to
    be replaced.

 6.5.2.1  Digital Pressure Indicator

 The digital pressure indicator operates using a small pressure sensor  and microprocessor to
 produce a reading in units of pressure measurement. These are typically hand-held units with
 excellent precision, resolution, and stability. Digital pressure indicators are required to be
 calibrated to a NIST-traceable standard on a regular basis (e.g., yearly) to ensure precision and
 proper operation. In most cases, these units must be returned to the vendor or sent to a
 metrology laboratory for this calibration.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2. 12. 6.0 Calibration Procedures                                January2016     Page16of18

 6.5.2.2  Fortin-Type Barometer Readings

The Fortin mercurial type of barometer works on fundamental principles of length and mass and
is therefore more accurate but also more difficult to read and correct than other types. By
comparison, the precision aneroid barometer is an evacuated capsule with a flexible bellows
coupled through mechanical, electrical, or optical linkage to an indicator. It is potentially less
accurate than the Fortin-type, but can be transported with less risk to the reliability of its
measurements (and it presents no hazard to personnel from mercury spills). The Fortin-type
barometer is best employed as a higher quality laboratory standard that is used to adjust and
certify an aneroid barometer in the laboratory. The certified aneroid barometer can then be taken
to the field and used to verify readouts from the transducer-type pressure sensor in the sampler.
The sampler sensor can be left in the sampler during the comparison because atmospheric
pressures are equivalent within a 100-meter horizontal distance and a 0.5-meter vertical distance.
If there is a discrepancy, the aneroid barometer should be recompared to the Fortin-type
barometer or other standard upon return to the laboratory. A discrepancy with the sampler sensor
could indicate an offset in the sampler's pressure sensor adjustment or the need for replacement
of the unit.

 6.5.2.3  Aneroid-Type Barometer

 The aneroid barometer is a mechanical instrument used for measuring barometric pressure using
 the principles of expansion and contraction to demonstrate changes in pressure. An aneroid
 barometer typically contains an evacuated cell or capsule that is prevented from collapsing by a
 strong spring. Small changes in pressure causes the cell to expand or contract driving
 mechanical levers to amplify the small movements and display them on the face of the
 barometer.

 6.6   Leak Checks

 Two types of leak checks are specified for use with PM2.5 samplers — an external leak check and
 an internal filter bypass leak check. The results for both of these procedures should be recorded
 in the site's logbook and/or on the monitoring organization's data forms. Sections 7.4.3 and
 7.4.4 describes the leak check procedures.

 6.7   Verification/Calibration Frequency

 Due to the dynamic nature of the PIVfo.s monitoring program, EPA has chosen to place
 measurement quality performance criteria, including the frequency and acceptance criteria of
 verifications and calibrations in one document, the PM2.5 Validation Template.  If guidance
 needs to be revised it can be revised in the validation template. This will allow one document to
 be revised if necessary and will avoid confusion when multiple documents have different
 acceptance criteria for the same check. Only guidance will be changed in the validation
 template; regulatory requirements will not be changed.

 The PM2.5 Validation Template can be found at:
 Checking to verify that a sampler's flow rate measurement system still reads within the


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures
                January 2016     Page 17 of 18
 acceptance limit of the previous calibration should be done according to the sampler
 manufacturer's operating instructions, should take place at the field site, and should be done at
 least annually or:

    • Following major electrical or mechanical maintenance, such as replacement of a circuit
        board or rebuilding of the sampling pump assembly;

    • When a routine flow rate verification indicates a variation from the set point (16.67 L/min)
        of more than 4 percent.

 If the annual verification fails, then recalibration and/or maintenance is required.  Before
 beginning a flow rate calibration, the operator should repeat the flow rate verification to be
 certain there were no leaks in the connections between the flow rate verification device and the
 sampler inlet,  and that the flow rate data were reduced and interpreted correctly. In addition, the
 operator should ensure that the collector's ambient temperature  and pressure measurement
 systems are responding accurately; if not, these systems must be recalibrated (or replaced and
 calibrated) before performing a flow rate calibration.

 The recommended intervals for calibration and verification of flow rate, temperature, and
 barometric pressure are summarized in Table 6-1.
 Table 6-1.  Calibration and Verification Check Intervals17
  Parameter

  Flow rate calibration (multi-point)
Recommended minimum interval

Upon failure of flow rate verification; also, on
installation and then annually or when out of
specifications
  Flow rate verification (single-point)

  Temperature calibration (multi-point; both
  ambient air inlet and filter temperature sensors)

  Temperature multi-point verification
  Temperature verification (single-point; ambient
  air inlet sensor, filter temperature sensor if
  convenient to do so)

  Pressure calibration
  Pressure verification (single-point)
Every 30 days +3 days and 12/year

Upon failure of multi-point verification
On installation, then annually or when out of
specifications

Every 30 days +3 days and 12/year
On installation, then annually or when out of
specifications

Every 30 days +3 days and 12/year	
17 See Validation Template on AMTIC for any revision to acceptance criteria in this table
http://www3.cpa. gov/ttn/amtic/qapolliitant.html
                                                 2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.6.0 Calibration Procedures
January 2016    Page 18 of 18
PM2.5 Calibration Work Sheet
Sampler Type
Field Scientist
Site ID:
Sampler Id:
Date:
Site Name:
Calibration Standard Information
Device
Temp.
Pressure
Flow

Make/Model



Serial Number



Last Verification of Standard



Calibration Data
Temperature Calibration
Device
Ambient Standard
Ambient Sampler
Ambient Filter
Cold Standard
Cold Sampler
Cold Filter
Ambient Standard
Ambient Sampler
Ambient Filter
Hot Standard
Hot Sampler
Hot Filter
Ambient Standard
Ambient Sampler
Ambient Filter
Value 1















Data Evaluation Sampler Temp
Ambient- Standard vs Sampler
Cold- Standard vs Sampler
Hot- Standard
vs Sampler
Average Difference
Value 2















Difference




Values















Value 4















Value 5















Data Evaluation Filter Temp
Ambient -Standard vs Filter
Cold- Standard vs Sampler
Hot- Standard vs Sampler
Average Difference
Ave















Difference




One-Point Temp Verification
Standard Value

Sampler Value Difference

Barometric Pressure
Low Point
Standard


Sampler

Diff.

Ambient
Standard

Sampler

Diff

High Point
Sampler Standard Diff

Flow Rate (acceptance Criteria + 4 % of standard)
Point #1
Standard

Sampler

% Diff.

Point #2
Standard

Sampler

%Diff

Point #3
Sampler Standard %Diff

Sampler flow rate check after calibration or verification (acceptance Criteria + 2 % of design value)
Sampler Value

Comments
Design Value

% Difference


 Figure 6.1 Example Calibration Worksheet
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                        January2016    Page1of17


 7.0   Field Operation


 7.1    Overview

 Sampling operations provided in this section are independent of any one type of commercially
 available PIVfo.s sampler. Because operational procedures may vary among sampler models,
 consult the manufacturer's operations manual before the sampler is put into service. Sampling
 procedure checks are summarized in Table 7-1.

 7.2   Activities to Perform Each Site Visit

 This section reviews the steps taken during a site visit to prepare for and complete a sample run,
 sample validation criteria, and proper handling of the filter/cassette assembly. Also, refer to the
 sampling procedures summarized in Section 3.4.

 Samplers used for NAAQS regulatory decision making purposes are required to operate in
 accordance with the EPA's national sampling schedule.  The yearly schedule is posted on the
 EPA's AMTIC website18.  Depending on the sampler's designation (i.e., primary or
 collocated), as well as its location, the sampler will operate on either a daily, l-in-3 day, l-in-6
 day, or l-in-12 day sampling schedule. The operator will prepare samples and schedule field
 activities based upon the required sampling frequency of the sampler.

 For those samplers that operate on an intermittent sampling schedule, the loss of a small number
 of samples in any calendar quarter (such as those lost by power outages or sampler
 malfunctions) can potentially  impact data completeness.  The monitoring organization can
 choose to schedule and run make-up samples, when necessary, in order to help the monitoring
 organization achieve the requisite number of quarterly samples.  Five make-up samples can be
 collected for any sampler each quarter.  The monitoring organization's make-up sample policy
 should be documented in its QAPP, and procedures for conducting make-up samples provided
 in the monitoring organization's SOP.  The operator should be aware of the make-up sample
 procedures and be prepared to schedule make-up samples, when needed.  In general, the
 operator should attempt to schedule a make-up sample as soon as possible following the missed
 sample, or schedule the make-up sample exactly one week from the missed sample, on the same
 calendar day.

 More  detailed EPA guidance on conducting make-up samples can be found on the AMTIC
 website19.
18

is ii«r.-/Ai,ii,i.,"i ^reyjov/ttii/ainl^^
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                         January 2016    Page 2 of 17

 7.2.1    Beginning a Run

 Care must be taken to ensure that the filter is clean and undamaged before it is installed in the
 sampler. The filter/cassette assembly should be kept in its protective container until installation;
 damaged filters must be placed in the protective container and returned to the weighing
 laboratory, where they will be examined and then discarded.

 1.   Annotate the following information on the sampler run data sheet, in the sampler logbook,
     and/or on the sample COC, as required by the monitoring organization's QAPP or SOP.
     The single-filter PM2.5 Field Data Sheet shown in Figure 7.1 is an example of such
     documentation.

       •    Date and time of sampler setup visit

       •    Site identification and location

       •    Sampler model, unique sample ID number (this may be the cassette and/or filter ID
           number, or some other tracking number)

       •    Scheduled sample start  date and time

       •    Unusual conditions that may affect samples (e.g., subjective evaluation of pollution
           on that day, construction activity, weather conditions)

       •    Setup operator's signature or initials.

       NOTE:  Figure 7.1 is an example data sheet for a single-filter sampler and assumes
       the data sheet will accompany the sample to the laboratory. Because sampler data are
       electronically downloaded and archived, table entry of data relative to the run start
       and end operating conditions may not need to be made, so long as the information is
       properly documented in electronic format and appropriately stored.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation
                                          January 2016     Page 3 of 17
  Table 7-1.  Sampling Procedure Checks20
       Procedure
    Frequency and
       method
     Requirements
       Action if
 requirements not met
   Filter installation
   (Never remove filter
   from cassette for
   inspection!)

   Sample validation
   and documentation
   Post-sample
   inspection
   External
   Leak check
   Flow rate checks
   Field blank check
Visually check filter.
Visually check each
sample and the keypad
display or downloaded
sampler data for
completeness.


Visually check filters
while in cassettes for
tears, missing pieces,
or leakage. Review
sampler ope ration.

Every flow rate
verification or
calibration, when
suspected, and before
and after separator
cleaning/maintenance.

Check flow rate at least
every 4 weeks or once
per month at each
sampler in the network.

At least one field  blank
available per weighing
session. Install, then
immediately remove
filter from sampler;
store in protective
container inside
sampler case. Or
preferably, install in idle
single-filter sampler for
24 hours, then remove
and process. For
sequential samplers,
install in unused
holder.
Filters must be uniquely
identified, tare-weighed,
undamaged, and in
cassette.

Record sampling date,
filter and sampler ID,
station location, flow
rates, sample time, and
unusual conditions on
data sheet.

There should be no
evidence of filter damage
or sampler malfunction.
Leak-check results must
be within parameters
specified by
manufacturer.
Indicated sampler flow
rate must be within ±4
percent of the measured
flow rate.

Rotate blanks from
sampler to sampler, so
all in network are
checked. Special
requirements for
sequential samplers:
should have one or more
field blanks with each
batch of filters to be
weighed.
Void the filter, and
install substitute
filter/cassette.
Complete or correct the
documentation. Cross
out invalid information
with a single line. Initial
and date changes.


Flag the sample as
questionable; correct
the cause of
malfunction.
Determine cause of
leak and correct.
Validate and/or
calibrate the sampler
flow rate.
Correct problems.
Recalibrate the sampler
if needed.
Reassess
filter-handling
techniques,
transport/storage
conditions, and sampler
cleanliness.
20 See Validation Template on AMTIC for any revision to acceptance criteria in this table
http://www3.cpa.gov/ttn/amtic/qapolli.itant.html
                                                  2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation
January 2016     Page 4 of 17
                                Sampler Field Data  Sheet
                                      Sampling Event Information
AQSSitelD POC:
Site Name
GPS Latitude
GPS Longitude
Para meter Check Device
Multi-Standard'
Temperature Standard
Barometric Pressure Standard
Flow Rate Standard
Time Checks OK?
Monitoring Site Criteria OK?
Make Model




PEP Setup Date
PEP Field Scientist
Primary SLT Sampler Serial No
PEPPQ200A Serial No.
Serial No.








Last Calibration Date




13 Yes D No (descnbe)
3 Yes H No (describe)
 •' Use this line for muin-nandard instruments fe.g.. BGI TriCai and DeitaCaii when used for ail ihree checks
Sampler Verification Checks (if performed) 2
Leak Check
2-Minute Interval
Bar. Pressure
Ambient
Temperature
Ambient Sensor
Filter Sensor
Criteria
Change < 5cmH:O
Criteria
± 10 mmHg
Criteria
±2-C
±2:C
Beginning P
cmH:0
Ret Standard
mmHg
Ref Standard
:c
°c
Ending P
cmHjO
Sampler
mmHg
Sampler
:C
°c
Date:
Verification OK?
3 Yes 3 No
Verification OK?
3 Yes 3 No
Verification OK?
3 Yes 3 rjo
3 Yes 3 No
Flow Rate Verification
Audit Standard
FR (Cai ) ChecK
Design Flow Rate
•Q" Check
Criteria
< 4% difference
Criteria (±4%)
16.00 it Fill out a ne\\' Field Data Sheet for the repiacemen t sampler Keep oldFDSfor record of trouble with sampler	
                                             Exposure Data
Filter Cassette ID
Elapsed Time (ET)
Total Volume (m-5)
Flow Rate (Lpm)
Start Date/Time
Stop Date/Time
Temperature (°C)
Bar Pressure (mm Hg)
Field Blank Cassette ID
Trip Blank Cassette ID
Companion Cassette ID*
Collocated Cassette ID(s)!



Cassette Retrieval Date/Time
Filter Integrity OK?

Q 167 Avg


Data Download OK?

Max: Min
Max Min:


3 Yes 3 No (describe)
cv
3 Yes 3 No (describe)
Avg
Avg
Sampler Flags':
Field Flags:


 •Make sure to add /ESI) flag in "Sampler Flags" if runtime is outside of1380-1500 minute range
 ' For P.\f-coarse sampling event. ifP\{-2.5 is routine filter type, then list the companion PM-10 filter cassette ID and vice versa
 ' For parking lot studies, all the IDs can be listed on one form. Be sure to indicate P.M cut point for each one.
 Notes:

   Figure 7.1 Example PM2.5 field data sheet.
                                                        2.12-PM2.5 Reference/Class I  Equivalent Methods

-------
2.12.7.0 Field Operation                                         January 2016     Page 5 of 17

 2.   Ensure the sampler is not operating. If the sampler is set to automatically begin operation,
     ensure that enough time is available to complete these setup procedures before it starts.

 3.   Open the filter holder assembly according to the manufacturer's instructions. Visually
     inspect the O-rings inside the filter holder to ascertain that they are present, secure and not
     cracked. Do not sample without these O-rings installed because the system will no longer
     be leak-free. Install the uniquely identified filter cassette containing the pre-weighed filter.
     Never remove the filter from the cassette. This is done only at the filter weighing facility.
     Reinstall the filter cassette holder and ensure that the fittings around the impactor housing
     and the filter assembly  are secure.

 4.   The sampler is now ready to sample. If the sampler is not already set to turn on
     automatically for the next sampling period (i.e.  midnight, for SLAMS), program the
     controls to do so. Check clock on sampler to ensure it's accurate  to within 1  minute of an
     atomic clock and set to local standard time (for SLAMS).

 5.   Visually inspect the monitoring site and its equipment to ensure that all sampling
     components are ready for the next run day(s). Note any changes in the site surroundings,
     especially dust-producing activities.

 6.   Visually inspect the records of the sampler. If it is time for the monthly/every 4 weeks
     check, measure and record independent measurements of ambient temperature and
     pressure, and ensure that the ambient temperature (i.e., inlet temperature) and pressure
     readings taken by the sampler are within 2.0 °C and 10 mmHg of the independent readings,
     respectively. Be certain that the independent temperature sensor (thermometer or thermistor
     probe) is located side-by-side with the sampler's ambient temperature sensor. This will
     require that the sensor be placed in the louvers of the sampler's radiation shield and kept
     out of direct sunlight. Also check the sampler's display for the filter temperature and ensure
     this value is reasonable compared to the ambient temperature display.

 7.2.2    Ending a Run

 1.   Visually inspect the sampler readouts to ensure that the sampler is operating properly.
     Sequential samplers require manipulation of a number of display screens to retrieve all
     data. Consult the operating manual. Also check the sampler for any other obvious
     problems, such as a full water collection jar. If problems are identified, describe them on
     the sample run data sheet and take corrective actions before starting another run. If the
     weather is bad,  provide a temporary shelter to facilitate data transfer and to protect exposed
     parts of the sampler. A  small work table may be useful.

 2.   Record the following information on the  sampler run data sheet, in the sampler logbook,
     and/or on the sample COC (see Figure 7.2),  as required by the monitoring organization's
     QAPP or SOPs.

   •   Date and sample pick-up time (Appendix L, Section 10.10)

   •   Stop time and total elapsed time of the  sample run
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                          January 2016    Page 6 of 17

   •   Average flow rate, coefficient of variation of the flow rate, and total volume sampled

   •   Average temperature and pressure measured by the sampler during the sampling event

   •   Conditions at the site or of the collector that may have affected the sample

   •   Any flags triggered by the sampler (e.g., power outage, flow rate variation)

   •   Explanations for questionable or voided samples

   •   Collecting operator's signature or initials

 3.   Download the runtime data (e.g., sampler filter, interval, and input data files) for the
     completed run using a laptop computer or other data transfer device that may be specified
     in the particular sampler's operating manual.

 4.   Carefully open the sampler's filter holder assembly according to the manufacturer's
     instructions. If the filter cassette comes apart or sticks to the upper housing during this
     process, close and gently reopen the assembly. Do not allow the filter to be shaken,
     dropped, or touched by any foreign object (fingers, rain, and so on). Visually examine the
     filter and cassette for damage or unusual appearance. Make notes and then immediately
     place the filter cassette inside an appropriately marked protective container (see Figure 4.2)
     for storage and later transport to the weighing laboratory.

 5.   Inspect the interior of the filter housing and the sampler itself. Note any abnormalities on
     the sampler run data sheet.

 6.   Conduct any scheduled maintenance activities. Refer to Section 8.2 for details on separator
     inspection, cleaning, and/or replacement.

 7.   If another sampling run will be performed, install a filter cassette according to the
     instructions in Section 7.2.1, above, and  the manufacturer's instructions.

 8.   Observe conditions around the monitoring site; note any activities that may affect filter
     particle loading (e.g., paving, mowing, fire) and record this information on the run data
     sheet. Retrieve equipment and applicable documentation. Secure the site.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation
                                                                  January 2016     Page 7 of 17
                 PM2.s  Sampling and  Chain-of-Custody Form
  PART I - WEIGHING LABORATORY
            Filter Weighing and Shipping Information from Weighing Lab or Shipping Log
Filter ID
Weighing Lab
Analyst/ Custodian
Shipment Date
Sent to (PE Org)
Date This Filter Must be Used bj
Filter Cassette ID
Cassette Type
Tare Weight Date
Tracking No.
Shipping Company
Q TB - Trip Blank




Return to:
  .Yormam, t>:e \\etgiiing laboratory completes Pert I, iseeps I cop\ ami semis 2 copies to ihe field office weft the unexposedfiller c/xserte
  PART n - FIELD OFFICE
    Date Received:
                                Received by
Location
    Package Condition: D Good  P Reject (Why?)
  If rejected, the fitter cassette should be returned so the weighing laboratory •with the next outgoing shipment
  PART m - FIELD SITE
                                     Sampling Event Information
              Arrival Date at Site
                                                         Field Scientist:
         Site Name & Description
            SLTPM-2.5Sanpier Make/Model:
                                                         Senal No
                   AQS Site ID
                                                         POC:
     Cither Operators or Observers
                                      Sampling Event Filter Data
    Sampling Date
                                        | Retneval Date
              Time
    Event Fitter Integrity   ~ OK  C. Reject (describe)
    Sam pie Type
    G RO - Routine          n FB - Field Blank (Associated RO Cassette ID..
    D CO - Collocated        D Expired Filter (not used)
    G TB - Tnp Blank (Record last RO Cassette ID used in this audit tnp:	
                                                                             D Other (d esc nbe)
    n Void (why?)
PM-2.5 Separator Type:  n WINS
                                     VSCC
  PART TV - FIELD FILTER SHIPPING TO WEIGHING LAB
Shipment Date
Shipped by
Tracking No
Affiliation
Shipping Destination
Shipping Company
  On completion of Pan tt-W, the field scientist keeps one copy and sends the top foriginalj copy to the laboratory with the filter.
  PART V -WEIGHING LABORATORY
Date Received
Shipment
Integrity OK?
D Yes D No
Received by
Max Temperature:
*C
Integrity Flag:
Cold Pack Condition: n Frozen D Cold
H Ambient
  The neighing laboratory will DATE-STAMP and attach the COCform to the receiving log-book, in which same info is recorded
  Notes:
  Figure 7.2 Example filter chain-of-custody
                                                     2.12-PM25 Reference/Class I  Equivalent Methods

-------
2.12.7.0 Field Operation                                         January 2016     Page 8 of 17

 7.2.3    Sample Validation

 Observations made by the site operator concerning the sampler's operation are important in
 deciding if a sample is valid. Never discard a filter/cassette. If sample validity is in question,
 operators are encouraged to use data qualifiers or flags and to record free-form notes on the data
 sheet, in a site logbook, or on computer data entry screens. The final decision for invalidation
 will be made after the information provided by the operator and the filter/cassette itself are
 examined at the office and laboratory. Tight control of sampler operation is important because
 too many invalid samples could cause an entire quarter's data set to be lost. The following
 criteria have been established to assist in the determination of data validity. CFR references for
 the acceptance criteria can be found in the validation template.

 1.  Timing

All samplers must operate for at least 23, but not more than 25, hours (1,380 to 1,500 minutes).
Samples  taken for less than 23 hours are invalid for the daily average, but can be used to
determine exceedances of the PM2.5 standard. Samplers are programmed to stop whenever the
filter becomes so heavily loaded that proper flow rate cannot be maintained.

 2  Flow Rates

The average flow rate for a 24-hour period must be within 5 percent of 16.67 L/min at actual
conditions. If this limit is exceeded, investigate the cause. Use the following criteria as a basis
for sample invalidation:

    •   Deviations in flow rate during sampling (due to mechanical failure) of more than 5
       percent from the design value require a flow rate verification check. Refer to Section
       6.3.1 for information. Remember to use a flow check filter, not one used for sampling. If
       the sampler's flow rate verification check indicates that the comparison did not meet the
       ±4 percent acceptance criterion, the sample is questionable and should be flagged for
       potential invalidation.

    •   If the sampler flow rate decreased because of heavy particulate loading on the filter, the
       sample should not be invalidated because the heavy loading may indicate an episodic
       situation that deserves study.

    •   Changes in flow rate calibration of more than 4 percent, as determined by a monthly field
       flow rate verification check, may cause invalidation of all samples collected since the last
       acceptable flow rate check. Make corrections or repairs, verify the flow rate, and
       recalibrate the sampler if required.

 3   Filter Quality

A sample collected on any filter that has obviously been damaged (i.e., torn, frayed, or has
pinholes) during the collection process should be invalidated. Return such a filter/cassette
assembly to the office or laboratory with notes from the operator.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                        January 2016     Page 9 of 17

 4.  Filter Temperature

Based on the data downloaded from the sampler for the run, the filter temperature must be no
greater than 5 °C above the ambient temperature for more than 30 consecutive minutes, based on
5-minute averages. Take corrective action if it is.

 7.2.4    Sample Handling

 7.2.4.1  Handling a Valid Sample

The sample must be removed from the collector within 177 hours (7 days 9 hours) of the end of
the run. Promptly deliver the filter cassette in its protective  container, accompanied by the
completed run data sheet and chain-of-custody, to the analytical laboratory. See Section 10.7 for
additional details on post-sampling filter inspection. Package and handle the samples to avoid
vibrations that may dislodge particles. Do not allow the filter surfaces to touch any of the  interior
surfaces of the protective container. Samples are to be transported in an insulated container
capable of maintaining a temperature of less than 25°C (77°F). 40 CFR Part 50, Appendix L,
Section 10.13 states, "During the period between filter retrieval from the sampler and the  start of
the conditioning period, the filter shall be maintained as cool as practical and continuously
protected from exposures to temperatures over 25°C to protect the integrity of the sample and
minimize loss of volatile components during transport and storage." The sample may be cooled
to 4°C by placing leak-proof ice substitutes inside the insulated container in order to keep the
samples cool, especially in the summer months when an operator may be in the field for several
hours. Do not allow the protective container to become wet. Place a min/max thermometer, a
liquid crystal temperature sensor strip, or other temperature recording device inside the shipping
container to verify that the sample(s) remained cool during transit to the laboratory. Please note
that samples are never to be  frozen (i.e., placed in a freezer). Handle field blank filters in the
same manner.

Depending on whether the sample is shipped or transported cold, or below the average ambient
air temperature  during sampling, the laboratory has from  10 to 30 days from the sample's end
date to determine the sample weight. Pursuant to 40 CFR Part 50, Appendix L, Section 8.3.6, the
post-sampling conditioning and weighing shall be completed within 240 hours (10 days) after the
end of the sample period, unless the filter sample is maintained at temperatures below the
average ambient temperature during sampling (or 4 °C or below for average sampling
temperatures less than 4 °C) during the time between retrieval from the sampler and the start of
the conditioning, in which case the period shall not exceed 30 days.

   NOTE: Unless one has some way to determine the maximum/minimum temperatures in
   the transport container during transport, there is no way to guarantee that the filters were
   not exposed to temperatures greater than the average sampling temperature. For example,
   samples collected in winter months or northern climates may actually warm above average
   ambient sampling temperatures during transport if ice substitutes are not used.  In this
   case, the laboratory would have 10 days to weigh filters.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                       January 2016    Page 10 of 17

7.2.4.2   Handling a Questionable Sample

  1.  Complete as much of the run data sheet as possible and explain any omissions.

  2.  Mark "questionable" on the run data sheet and record comments on the data sheet and/or in
     the site logbook.

  3.  Do not discard the filter.

  4.  Deliver the filter cassette in its container (as if it were a valid sample) and the run data
     sheet to the analytical laboratory. If your quality assurance (QA) program calls for it, notify
     the appropriate person of the potentially invalid sample.

     NOTE: Monitoring organizations may be using contract laboratories for filter analysis.
     There should be some contractual agreement that the contract laboratory will immediately
     contact the monitoring organization of any sample of questionable quality. The monitoring
     organization, not the contract laboratory, should make the decision whether or not to
     invalidate the questionable sample(s).

 7.3   Activities to Perform Every Five Operating Days

 7.3.1    Impactor Well  Cleaning

 Cleaning of the impactor well is very important. It is necessary to prevent the sampler from
 operating with an overloaded PM2.5  impactor. When operating in an area with higher PMio
 concentrations, more frequent cleaning of the well may be necessary. Refer to Section 8.2.2 for
 impactor maintenance guidance.

 In the case of sequential samplers in use every day, sampling may be interrupted for up to 1
 hour without invalidating the day's sample. Exercise care not to contaminate the filter during
 the well cleaning or replacement process.

 7.4   Activities to Perform Every Four Weeks

 The following activities are performed every thirty days:

    •  VSCC™ Cleaning (if using VSCC™ separator)
    •  External leak check
    •  Temperature QC  check
    •  Barometric pressure QC Check
    •  Flow Rate verification check

 There are other maintenance activities that are performed every 30 days that are discussed in
 Section 8.0.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                       January2016    Page11of17

 7.4.1    VSCC™ Cleaning

 Refer to Section 8.3.3 for details on VSCC™ cleaning. It is highlighted in this section due to its
 importance in achieving data quality.
 7.4.2   Field Quality Control Equipment and Standards

 The following equipment is required for field calibration checks of temperature, pressure, and
 flow rate.

    •   A thermometer or temperature sensing device readable to the nearest 0.1 °C, capable of
        accurately measuring temperature to the nearest ±1 °C, and referenced to a NIST
        standard or an American Society for Testing and Materials (ASTM) thermometer within
        ±0.5 °C at least annually.

    •   A barometer capable of accurately measuring ambient barometric pressure to the nearest
        ±1 mmHg and referenced to a NIST or ASTM barometer within ±5 mmHg at least
        annually.

    •   An orifice device (or other acceptable flow measurement device) and calibration
        relationship.

    •   The sampler's calibration information (an equation).

    •   A clean flow check filter, installed in a cassette.

    •   A flow check data  sheet (such as the example given in Figure 7.3) and/or the sampler's
        logbook.

 7.4.3   External Leak  Check

 An external leak check will test the sampler's entire sample train - from the inlet to the pump -
 for a leak.

 Before conducting filter temperature and flow rate verifications, perform an external sampler
 leak check first.  This is important in order to determine the "as found" condition of the
 sampler, and to ensure no leaks are present. Similarly, it is important to conduct an external
 leak check prior to conducting any separator maintenance, or any procedure which would result
 in the separator being removed from the sampler (such as accessing the filter temperature sensor
 in some sampler models).  The removal of the WINS impactor or VSCC™ from the PM2.5
 sampler will  break the seal within the sample train, which could result in a leak if not properly
 reassembled. Therefore, the operator should determine the "as found" leak status, and then
 follow any separator maintenance (or similar) with a second leak check, in order to verify the
 sampler leak status "as left".
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation
                                                            January 2016    Page 12 of 17
 The following describes a generalized procedure for conducting an external leak check.

 1.   Insert the clean filter (designated the "leak check filter") into the sampler filter holder using
     the appropriate procedure. Leak check filters should never be used for subsequent
     sampling. The same filter may be used for the leak check that was used for the flow rate
     verification check.

 2.   Remove the inlet and install the flow adapter (sometimes referred to as the flow shut-off
     valve) on the top of the downtube. Close the flow adapter valve to prevent air flow.

 3.   Perform the leak check procedure according to the specific instructions in the
     manufacturer's operating manual.
6.
                                       CAUTION

           Open the flow adapter's valve slowly to prevent a sudden rush of air
           into the system, which can cause oil to splash from the impactor well
           onto the filter and other components. Opening the valve too quickly can
           also rupture the leak check filter.
 4.   Upon completion of a leak check, the sampler's pump may turn off automatically. If not,
     turn off the pump. If the sampler leak check result is within acceptable limits, as detailed in
     the operating manual, then the sampler is operating properly and you may proceed with the
     temperature, pressure, and flow rate checks.  If the leak limit is exceeded, investigate the
     cause and make corrections.  For example, check the O-rings on the flow adapter for nicks,
     wear, and excessive flattening. Also, check the outside of the intake tube (i.e., downtube),
     where the leak check device sits, for deep scratches or dirt that could allow leakage. Check
     that the filter holder is well-seated. Rectify  any problems and repeat the leak check. An
     internal leak check may be necessary (see Section 7.4.4.).  Document the sampler logbook
     and/or any data forms required by the monitoring organization with details of the leak test
     procedure and any necessary corrective actions. If necessary, recalibrate the sampler before
     sampling is resumed.
     Remove the flow adapter and replace the inlet.

     Remove the leak check filter from the filter holder. This filter is never reused for sampling
     but, if carefully handled, may be used several times as a leak check filter.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation
          January 2016    Page 13 of 17
                      PM2.5 ROUTINE QUALITY CONTROL CHECKS
                                 & MAINTENANCE FORM
                     SITE
           INSTRUMENT
       AQS Site No.:
          Location:
          Operator:
 Manufacturer:
   Model No.:
  Inv./Ser. No.:
      WINS IMPACTOR or CYCLONE SERVICING
      INSTRUMENT CLEANING
• WINS OvSCC Rubber Seals
WINS Impactor/VSCC Change Date: Clean Inlet Head
New Impactor/VSCC No.: Clean Air Screens
Clean Downtube
• Oraned
O Replaced
Yes ONO
©Yes O"*>
©Yes
O"*1
                                       LEAK CHECKS
                  EXTERNAL
            INTERNAL
Date Performed
Value
If FA ll.f.fi (value prcal
operator17
IfYF.s , Corrective
Action Taken

mm Hg
rtban 2? mm Hg), was problem resolved by
Q Yes Q No



Date Performed
Value
If f All Kn (value grea((
operator7
(fYES .Correclive
Ac lion Taken

mm Hg
rthan HOmm Hg),was problem resolved by
fj Yes fj No


                  TEMPERATURE CHECKS
                                                       It AKOMETKIC PRESSURE CHECK

Actual Temp
Sampler Temp
± 2 Diff.
Ambient



Filter



Compartment








Actual Pressure (mm Hg):
Sampler Pressure (mm Hg):
± 10mm Difference (mm Hg):
H.OWRATE CHECK
Flow Device
Inventory #
Cert Date
Actual Flow
Sampler Flow
± 4% %d
Target Flow 16.67
± 5% to set %d
CALIBRATIONS NEEDED?
fJYES* D NO
*lfy?*, docMmrnt i'asH'alibralian Kesulti Xfclian
l>OST-CALIItRATI(>N RESULTS

Actual Temp
Sampler Temp
± 2 DitT.
Ambient



Filter



Compartment



'Flow recalibrabon required AFTBt Temperature or Pressure recalibration'
Barometric
Pressure
Actual

Sampler

± 10mm Diff.

FLOWRATE POST-C:AI,IBRATION
Actual Flow:
Sampler Flow:
± 2% %d :



1 ;
-------
2.12.7.0 Field Operation                                       January 2016     Page 14 of 17

 7.4.4  Internal Filter Bypass Leak Check

 Performing an internal leak check is only necessary when an external leak check fails.   An
 internal leak check test will assist the operator in isolating the location of the leak within the
 sampler.

 The suggested technique for implementing an internal leak check follows.
 1.   Install a flow-impervious membrane material in the filter cassette, either with or without a
     filter, as appropriate, which effectively prevents air flow through the filter.

 2.   Remove the inlet and install the flow adapter (sometimes referred to as the flow shut-off
     valve) on the top of the downtube. Close the flow adapter valve to prevent air flow.

 3.   Perform the internal leak check procedure according to the specific instructions in the
     manufacturer's operating manual. If the internal leak check passes, then the sampler
     components below the filter are leak-free, which means that the leak is located somewhere
     between the PMio inlet head and the filter. If the internal leak check fails, the leak is
     located below the filter.

 4.   Perform requisite corrective actions.  Document all troubleshooting activities and
     corrections in the sampler logbook and/or on any data forms required by the monitoring
     organization.

 5.   After corrections are completed, conduct another external leak check to ensure the source
     of the leak has been mitigated and the sampler is leak-free.   If necessary, recalibrate the
     sampler before sampling is resumed.   Document the results of the final leak test in the
     sampler logbook and on any necessary data forms.

 6.   Remove the flow adapter and replace the inlet. Remove the leak check filter from the
     filter holder.
 7.4.5   Procedure for Temperature Verification Check
PM2.5 samplers have at least two temperature sensors - an ambient temperature sensor and a
filter temperature sensor.  Some samplers will have an additional compartment temperature
sensor as well.  All temperature sensors should be routinely verified.  The acceptance criterion
should be ±2 °C for all temperature sensors as compared to the verification/certification standard.

Due to variations in sampler models, a procedure for accessing and verifying the filter and/or
compartment temperature sensors is not described here. Please consult the sampler user manual
for information on how to test these sensors.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                        January 2016     Page 15 of 17

The following describes steps to perform a single-point temperature verification of the ambient
temperature sensor only.

       1.  Allow adequate time for the field check temperature standard to reach temperature
          equilibrium with its surroundings. This may take 15 minutes or more. The site
          operator should use his or her best judgment to ensure the temperature sensor has
          equilibrated to ambient conditions.

       2.  The sampler's Main screen should be displayed.

       3.  Place the field check temperature standard between the slats of the gill screen
          (radiation shield) housing the sampler's ambient temperature sensor. Wait until the
          temperature check standard reading is stable, and then compare it to the ambient
          temperature reading displayed on the sampler's Main screen. If the temperatures
          agree within ± 2°C, then the ambient temperature sensor response is acceptable. If
          not, proceed to Step 5.

       4.  Record the ambient temperature information on the field data sheet.

       5.  If the readings are  outside acceptance criteria, wait longer (10 to 15 minutes) for
          temperature equilibration to occur and repeat the procedure. If  the readings still do
          not agree, verify that the problem is not with the standard by using another
          verification device, if available.

       6.  If after troubleshooting the readings  are still outside the acceptance range, record the
          temperature information on the field data sheet and take corrective action.


 7.4.6    Procedure for Barometric Pressure Verification Check

A single-point barometric pressure verification is described in the following steps.

       1.  Allow adequate time for the barometric pressure verification device (standard) to
          reach equilibrium with its surroundings. This may take 15 minutes or longer. The site
          operator should use his or her best judgment to ensure that the  standard is equilibrated
          to ambient conditions.

       2.  The sampler's Main screen should be displayed.

       3.  Place the pressure  standard in close proximity (< 0.5 meters) to the sampler's pressure
          sensor. Wait until the standard's reading is stable, and then compare it to the ambient
          pressure reading displayed on the sampler's Main screen.

       4.  Record the pressure readings from the sampler and the  standard on the field data
          sheet.

       5.  If the two readings are <10 mmHg of each other, the verification of the pressure
          sensor is satisfactory.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                        January 2016    Page 16 of 17

       6.  If the deviation is >10 mmHg, check the barometric pressure using a backup
          verification device (standard), if available. If the results are similar to those of the
          first check, then the sampler's pressure measurement system  may be damaged and
          should be serviced. A calibration procedure should be performed at a later time (see
          Section 6.5).


 7.4.7    Procedure for Field Flow Rate Verification Check

 A flow rate verification check of the sampler flow rate is required every 30 days. The results of
 all valid flow rate verifications are to be reported to AQS. Control charts presenting flow rate
 verification check data (indicated vs. measured) should be maintained. These charts provide a
 reference of instrument flow rate drift patterns and indicate when flow limits (±4 percent) have
 been exceeded. The field check is made by installing a measuring device (which is traceable to
 NIST and calibrated within the range of the flow rate) on the inlet of the sampler. Calibration
 procedures for the samplers flow rate measurement device are found in Section 6.3.4.

The following flow rate verification procedure is independent of any device. A variety of transfer
standards may be used with this procedure; however, the necessary apparatus and subsequent
calculations to determine the sampler's flow rate will vary.

 1.   Ensure that the instrument is operational and in working condition (i.e., not in Error mode).
     If the sampler is programmed to collect samples on a daily basis, the operator will have to
     stop the sample run in progress before proceeding with the flow rate verification procedure.

 2.   Insert the clean filter (designated the "flow check filter") into the sampler filter holder.
     Flow check filters should never be used for subsequent sampling but, if carefully handled,
     may be used several times as a flow check filter.

 3.   Read and record the following information on a sampler flow check data  sheet, such as the
     example given in Figure 7.3.

    •   Ambient temperature (To), °C (as indicated by the sampler's display)

    •   Ambient temperature (To), °C (as indicated by the field check thermometer or thermistor)

    •   Ambient barometric pressure (Pa), mmHg (as indicated by the sampler's display)

    •   Ambient barometric pressure (Pa), mmHg (as indicated by the field check barometer)

    •   Sampler ID number and model

    •   Flow verification check device (i.e., transfer standard) serial number and calibration
       relationships (if appropriate)

    •   Date, location, and operator's signature
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.7.0 Field Operation                                       January 2016    Page 17 of 17

 4.   Remove the inlet from the sampler downtube and replace it with the flow verification
     check device. Allow time for the flow verification check device to reach ambient
     temperature before use.

 5.   Begin the flow rate verification procedure. Allow the sampler to generate flow for at least 5
     minutes. Afterwards, check the display on the sampler and the flow verification check
     device to assess stability. The sampler is considered stable if the temperature, barometric
     pressure, and flow rate values are not fluctuating more than +0.1°C, +1 mmHg, and +0.1
     L/min, respectively, over a 60-second period. Record the flow rate measured by the flow
     verification check device and the flow rate indicated by the sampler.

 6.   Stop generating flow. Remove the flow verification check device and replace the inlet.

 7.   Remove the flow verification check filter from the filter holder

 8.   Using the information from Step 5 above, calculate the percentage difference as:

               % difference =  [(indicatedavg - measuredavg)/measuredavg\ • 100%

 9.   If the sampler flow rate is within 96 to 104 percent of the measured flow rate (at actual
     conditions), and if the sampler flow rate is within 95  to 105  percent of the design flow rate
     of 16.67 L/min, the sampler is operating properly. If either limit is exceeded, repeat the
     leak check procedure, as stated in Section 7.4.3. After doing so, investigate and correct any
     malfunction. If necessary, troubleshoot the sampler further, recheck the flow, and, if
     necessary, recalibrate before sampling is resumed.

 10.  Set up the sampler for the next sampling period according to the operating procedure in
     Section 7.2.1.

 11.  Have a supervisor or QC reviewer verify the information and data on the sampler flow
     check data sheet. The supervisor/QC reviewer should sign and date the data sheet after the
     review.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance                                     January 2016    Page 1 of 6


 8.0   Sampler Maintenance


 8.1   Overview

 Preventive maintenance is defined as a program of planned actions aimed at preventing failure
 of monitoring and analytical systems. The overall objective of a routine preventive maintenance
 program is to increase measurement system reliability and to provide for more complete data
 acquisition. This section addresses maintenance associated with reference or Class I equivalent
 PM2.5 monitoring methods only. Maintenance of the laboratory microbalance and support
 equipment is not covered. Laboratories often employ a balance service representative who
 services balances on a 6-month or annual basis.

 Some sampler maintenance can be performed at the field site. Major maintenance of the sampler
 or components, such as the pump, can be performed more conveniently when the equipment is
 brought to a laboratory or maintenance facility, provided time and labor are available to move
 the equipment.

 A maintenance schedule should be established for each sampler and systematic records should
 be kept as scheduled and unscheduled maintenance occurs. Files should reflect the history of
 maintenance, including all replacement parts, suppliers,  and cost expenditures and should
 include an inventory of on-hand spare equipment for each sampler.

 Recommended supplies for all maintenance activities include an alcohol-based general-purpose
 cleaner that leaves no residue, cotton-tipped applicators  (such as a Q-tip), a small soft-bristle
 brush, paper towels, distilled water, and miscellaneous hand tools. Additional supplies may be
 necessary for specific procedures, as indicated below. Specifications for many of the supplies
 are given in Section 4.0, "Procurement of Equipment and Supplies," of this document. A
 compressed-air source is also recommended but not required.

 All activities recommended below are in addition to those specified in the operating manual
 specific to the sampler and are not intended to replace those activities. Maintenance activities
 are summarized in Table 8-1. Site operators and/or maintenance personnel should review Table
 8-1 and develop appropriate schedules and documentation in their monitoring organization's
 QAPP to ensure the activities are met and reported. Maintenance activities should be included
 in the monitoring organization's standard operating procedures as  well.

 NOTE: Always remove the impactor and filter cassette  prior to removing the inlet downtube or
 cleaning any of the instrument parts upstream of these items. Such activities could dislodge dirt,
 oil, grease, or other materials that could deposit into the  impactor well  or onto the sampling
 filter. Temporarily store the impactor and filter cassette in a clean, dry location, away from
 contaminating materials (dust, dirt, rain,  and so on) and  direct sunlight.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance                                    January 2016    Page 2 of 6
  Table 8-1    Summary of PIVb.s Sampler Maintenance Items	
	Frequency3	Maintenance itemb	
 Every 5 sampling days             i.  Service water collector bottle (jar).
                                  2.  Clean or change-out impactor well.

 Monthly                          1.  Clean sampler inlet surfaces.
                                  2.  Clean impactor housing and impactor jet surfaces. Examine
                                     O-rings.
                                  3.  Clean interior of sampler case.
                                  4.  Check sampler clock accuracy.
                                  5.  Check condition of sample transport containers.
                                  6.  Clean VSCC™

 Quarterly (every 3 months)          1-  Inspect O-rings of inlet. Remove and lightly coat them with
                                     vacuum grease.
                                  2.  Clean sampler downtube.
                                  3.  Inspect and service water seal gasket where downtube
                                     enters sampler case.
                                  4.  Inspect and service O-rings of impactor assembly.
                                  5.  Inspect and service vacuum tubing, tube fittings, and other
                                     connections to pump and electrical components.
                                  6.  Inspect and service cooling air intake filter and fans.

 Other periodic maintenance         1.  As recommended by the manufacturer's sampler manual
 aFrequency may vary depending on climate, amount of particulate matter in the air, weather, and so on.
 bRemove impactor and filter cassette before servicing any upstream sampler components.
 8.2    Five-Day Maintenance Procedures

 8.2.1    Water Col lector

 At least every 5 sampling days, inspect the water collector bottle (jar). Remove accumulated
 water, clean the interior of the bottle, inspect the seals, and reinstall the bottle in the holder.

 8.2.2    WINS Impactor Well

 Impactor wells may be cleaned at the field site.  Take care to thoroughly clean the well and not
 to introduce contaminants. Others may prefer to have spare wells that can be cleaned in a
 laboratory or other indoor location. The well should be transported to the sampling site in an
 upright position. During transport, protect the well from excessive vibration, precipitation,
 direct sunlight, and other harmful environments. Likewise, the used well should be protected in
 transit back to the laboratory. It is recommended that wells be numbered or otherwise identified
 and records kept of their use.

At least every 5 sampling days, disassemble and clean the WINS impactor well (refer to Figure
8.2). If spare wells are available, cleaning can be done in the laboratory. Separate the upper and
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance
                                                                January 2016    Page 3 of 6
lower portions of the well, remove the used filter from the well, wipe the two halves of the well
clean with lint-free laboratory tissues or cloth, install a new borosilicate glass filter in the well's
bottom, and add 1 ± 0.1 mL of impactor oil. Check to be sure the oil covers the filter uniformly
and there are no air bubbles beneath the filter. Reassemble the well and place it in the impactor.
To reduce preparation time, several spare impactor wells should be prepared at one time and
stored in a clean, dust-free container until needed.
8.3     Monthly Maintenance Procedures

 8.3.1     Inlet                                  /
 Figure 8.1 shows a disassembled
 sampling inlet. The inlet seals to
 the downtube with two O-rings.

 If the water collector attached to
 the side of the inlet is glass, care
 should be taken during handling
 not to crack or break it because
 the sampler will not maintain an
 adequate vacuum during
 operation. The glass water
 collector may be either replaced
 with a plastic jar or wrapped with
 insulating tape to minimize the
 chance of accidental breakage.

 To dismantle and clean the
 sampler inlet:
                                               8
                                      Spacer
-HI
                                      Screw
                                                                         Top Plato with
                                                                         Deflector Cono
                                                                         Screen
                              Lower Plate with
                              Rain Deflector
                                                                         Acceleration Assembly
                                                                         O-ring
                                  Water Collector
                                                                        Collector Assembly
                                                                        O-ring Connection to
                                                                        Downtube
                                      Figure 8.1 Cross-sectional view of a PM10 sampler inlet head.
    •   Mark each assembly point
        of the sampler inlet with a
        pen or pencil to provide
        "match marks" during reassembly. Critical assembly points are already "keyed."
    •   Disassemble the sample inlet unit according to Figure 8.1, taking care to retain all the
        parts. NOTE: If the assembly screws appear frozen, the application of penetrating oil or
        commercial lubricant will make removal easier.

    •   Using a soft brush and/or cloth, lightly scrub all  interior surfaces and the bug screen with
        distilled water and/or general-purpose cleaner. Pay particular attention to small openings
        and crevices.  Cotton-tipped applicators and/or a small soft brush are most helpful in
        these areas. Using laboratory tissue and cotton-tipped applicators moistened with
        distilled water, wipe all surfaces to remove any remaining deposits. Completely dry all
        components.

    •   Check all the O-rings for distortion, cracks, fraying,  lack of lubricating grease, or other
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance
             January 2016     Page 4 of 6
        problems. Apply a thin coating of vacuum grease or replace the O-rings as necessary.

    •   Reassemble the unit in accordance with the previously scribed match marks. Take
        particular care to ensure that all O-ring seals are properly sealed and that all screws are
        uniformly tightened.
 8.3.2    WINS Impactor Housing and Well
 Figure 8.2 shows atypical disassembled
 impactor unit. With the impactor assembly open,
 inspect the interior of the impactor housing, both
 above and below the impactor well. These areas
 should be clean and dry. If necessary, clean the
 areas with a lint-free wipe. On a monthly basis,
 clean the interior of the impactor jet using a lint-
 free pipe cleaner or similar tool.

 Check the O-rings for distortion, cracks, fraying,
 lack of a light coating of vacuum grease, or
 other problems; replace as necessary.

8.3.3      Cleaning of VSCC™ type PM2.5
Separators (VSCC™ and similar devices)

 Cleaning of the VSCC™ well is strongly
 recommended after 30 days of operation. Also,
 empty  any water from the water trap (jar), clean
 the sampler interior, inspect the seals, and
 reinstall the trap. Cleaning the cyclone is
 necessary to prevent the sampler from operating
 with an overloaded PM2.5 separator. When
 operating in an area with higher PMio
 concentrations, more frequent cleaning of the
 well may be necessary. Remove the VSCC™
 from its installed position in the instrument.
         \
                  impactor Housing, Upper

                  	  O-ring

                   impactor Well, Upper
                   O-ring
             __—--• Impactor Well, Lower
                  Impactor Housing, Lower
         .-  —	-  O-ring
         ..	_.  O-ring
 SL
                  Filter Housing, Upper
               —	—  O-ring
            ,,_	-  Filter Cassette
Figure 8.2 Cross sectional view of PM2.5
impactor well and filter holder
       1.  Pull off the side transfer tube. If it is too tight to remove by hand, pry it off with a
          rigid plastic lever. Care should be taken to not damage the two O-ring seals.
       2.  Remove the top cap and grit pot by unscrewing.
       3.  Wet a low-lint wipe with water and  remove all visible deposits. These are most likely
          to be found at the bottom of the cone and inside the grit pot.
       4.  Inspect all O-rings for shape and integrity. If at all suspect, replace.  Lubricate all O-
          rings with light grease. It is important to well lubricate the transfer tube to avoid
          difficult disassembly.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance                                    January 2016    Page 5 of 6

       5.  Assemble in reverse order and reinstall.
       6.  Perform a leak check according to manufacturer's operating manual specifications.

 Separators may be cleaned at the field site. Others may prefer to have spare VSCCs™ that can
 be cleaned in a laboratory or other indoor location.


 8.3.4    Other

 Wipe down the interior of the sampler's case to remove bugs, dirt, and/or water deposits that
 may have collected inside the case. This procedure may be required more frequently during
 summer months. Inspect the cooling air intake filter and clean, if necessary.

 Check the sampler's internal clock against a timepiece known to be accurate within
 1 minute/month. It is recommended that a cell phone (with an atomic clock) be used for this
 purpose. Record the value and note whether the sampler's clock has gained or lost time since
 the previous month's time check. The sampler clock should be set to local standard time (for
 SLAMS).

 Check the maintenance (flow/leak check) filter cassettes and the backing screens for fractures,
 cracks, buckling, evidence of wear, or contamination.  Clean or replace as necessary. Conduct
 leak test(s) after reassembly.

 8.4    Quarterly Maintenance

 The following activities should be performed once every 3 months at the time of the regular
 monthly maintenance.

 8.4.1    Downtube

 Inspect the outer and inner surfaces of the tip (closest to the sampler inlet) of the downtube, and
 remove any particulate deposits using isopropyl rubbing alcohol or water and a soft bristle brush
 or wads of lint-free laboratory wipes or cloth. Dry the downtube completely before
 reinstallation.

 Inspect the rubber water seal gasket (located at the point where the downtube enters the
 sampling case) for cracks or other evidence of leakage. Lubricate or replace the gasket as
 necessary, according to the sampler's operating manual.

 8.4.2    Impactor

    •  Remove the O-rings in the impactor assembly  and condition them with a light coating of
       vacuum grease.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.8.0 Sampler Maintenance                                    January 2016    Page 6 of 6

 8.4.3    Other

    •  Examine the vacuum tubing for crimps, cracks, breaks, water, or dirt and replace as
       necessary.

    •  Examine the tubing fittings to ensure proper seating and tightness.

    •  Clean or replace the cooling air intake filter and fans as required.

 8.5   Other Periodic Maintenance

 As recommended by the manufacturer's sampler operating instruction manual.

 8.6   Refurbishing PIVh.s Samplers

PM2.5 samplers that have been operated in the field for extended periods may require major
repairs or complete refurbishment. In these cases, the manufacturer's manual must be referred to
before work is undertaken. A sampler that has been subject to major repairs or refurbishment
must be leak-checked and calibrated prior to sample collection. Such a sampler should be treated
as if it were a new purchase and all verification checks should be performed  prior to use.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                     January 2016    Page 1 of 14


 9.0    Gravimetric Laboratory Design and Set-Up

 9.1    Overview

 The quality of data from the PM2.5 sampling program depends on several factors. A primary
 consideration is the set-up of equipment and resources utilized within the gravimetric laboratory
 itself.  This section offers recommendations on designing the gravimetric laboratory (filter
 conditioning environment) in a manner which will optimize its performance, as well as provides
 recommendations on associated maintenance activities.

 Filter weighing activities are discussed in Section 10.

    NOTE: Monitoring organizations may contract pre-and post-weighing activities. Since
    contract laboratories may follow internally developed QAPPs and SOPs, it is the
    responsibility of the monitoring  organization to ensure those QAPPs/SOPs meet PM2.5
    requirements and are sufficient for their monitoring program. The monitoring
    organization should review the laboratory's QAPPs/SOPs on a routine basis, maintain
    copies of the documents, and request to be notified of any revisions. It is also the
    responsibility of the monitoring  organization to evaluate contractor compliance with the
    QAPP/SOP requirements through technical systems audits or other oversight activities.
    If the monitoring organization desires the contract laboratory to adhere to the monitoring
    organization's specific QAPP/SOPs, it is the responsibility of the monitoring
    organization to develop a contract with the laboratory that specifies such requirements.

 9.2  Weighing Room

 The gravimetric analysis of PM2.5 filters should occur in a climate-controlled, draft-free room or
 laboratory (i.e., filter conditioning environment).   It is imperative that the weighing room be
 capable of meeting the filter conditioning specifications that are detailed in 40 CFRPart 50,
 Appendix L, Section 8.2.  To that end, the room selected must not be subject to large
 temperature or humidity variations.  Selecting an interior room (i.e., a room that is not adjacent
 to an exterior wall) is strongly recommended;  moreover, a room without windows is preferred.
 Ensuring that the room is well insulated will facilitate control of temperature and relative
 humidity fluctuations.   Ideally, a small interior room should be selected because it is easier and
 more economical to control the environment in a smaller space.

 The filter conditioning environment should be maintained as a semi-clean room. The weighing
 room should be constructed  such that it is constantly under positive pressure, in order to avoid
 entrainment of foreign materials contaminating the room. Positive pressure can be verified by
 holding a piece of paper against the  outside door of the weighing room.  The paper should blow
 slightly outward, away from the door, if positive pressure exists. In order to further decrease the
 opportunity for contaminates entering the room, ingress to and egress from the weighing room
 should be minimized.

 Heating and air conditioning in the filter conditioning environment/weighing room should be
 maintained 24 hours a day, including weekends.  The monitoring organization  should be


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                      January 2016     Page 2 of 14

 cognizant of the expected temperature and humidity ranges in their local area throughout the
 year and utilize an HVAC system that can accommodate those ranges.  Employing a high-
 efficiency particulate air [HEPA]-filtered air supply system on the inlet air system will
 minimize human traffic effects and will help stabilize the conditions of the weighing
 environment. The monitoring organization should have a service contract in place to ensure
 proper maintenance of the HVAC, as well to provide vendor services when/if repairs are
 necessary. Re-heat systems also work well.

 Section 9.3 provides recommendations regarding the proper placement and installation of the
 microbalance within the filter conditioning environment.

 The weighing room should be well lit, in order to facilitate filter inspection and provide an
 efficient work space for the analyst.  The use of a light box is encouraged. Additionally, a
 space within the laboratory should be designated specifically for filter conditioning.   In
 accordance with 40 CFR Part 50, Appendix L, sample filters must condition (equilibrate) in the
 same environment in which the microbalance is placed.  The use of a conditioning rack with
 equilibration trays, shelves, or similar apparatus should be utilized for filter conditioning,  so
 long as the  apparatus is clean, static-neutralized (see Section 9.6), and allows for air flow over
 the filters during equilibration.

Preferably, the filter conditioning environment/weighing room should be dedicated to the use of
balances.  The monitoring organization is encouraged to limit the activities within the weigh
room to PM2.5 sample analysis only, in order to minimize the potential for cross-contamination
with other particulate types.

Dust contamination should be kept to a minimum. Routine cleaning regimes should be
developed and scheduled.  The use of semi-wet methods to clean (such as anti-static solutions
and damp lint-free cloths) is encouraged.  Analysts should avoid using brooms or other devices
which could create dust.  Examples of cleaning regimes for the weighing room may include:

Daily: Wipe down work surfaces used to perform laboratory activities with a lint-free  wipe that
has been dampened with alcohol-based,  anti-static solution.

Monthly:  Wipe down all surfaces, including the filter conditioning shelves, with a clean lint-
free wipe that has been dampened with alcohol-based, anti-static solution.   Mop the floor with a
damp clean room mop.

Yearly: Wipe down all surfaces, including walls, ceiling tiles, equipment, and work surfaces
with a clean, lint-free wipe that has been dampened with an alcohol-based,  anti-static solution.
Mop the floor with a damp clean room mop. Ceiling tiles, if removable, should be removed and
thoroughly cleaned.

NOTE: If filters are in petri-slides during these suggested monthly and annual cleaning regimes,
place the lids on all filters that are exposed in the conditioning environment to prevent
contamination.  It is further recommended that these cleaning activities be  completed during
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                     January 2016    Page 3 of 14

times when minimal weighing activity is expected.  All cleaning activities should be
documented in the laboratory logbook.

 Other clean room measures, such as installing sticky floor covering on the entrance(s) to the
 weighing area, should also be employed.

9.3     Microbalance

 Gravimetric analysis of the filters is performed using a microbalance with a readability of 0.001
 mg (1 ug) and a repeatability of 1 ug as listed in the microbalance's performance specifications
 (see Section 4.3.6). Because of the greater sensitivity needed for measuring microgram-range
 weights or weight differences, microbalances are vulnerable to relatively small changes in
 physical environmental conditions, such as vibration, electrostatic charge buildup, temperature,
 or relative humidity (RH). Features to offset the effect of these variables on the measurements
 may be integrated into the design of the microbalances, or they may be offered as options.

 The microbalance must be located in the same controlled environment in which the filters are
 conditioned. The microbalance used in the weighing process should be identified by a unique
 balance number. Make sure that the microbalance has been  externally calibrated/certified (at
 least annually) and maintained according to the manufacturer's recommendations. If it is out of
 calibration, the microbalance should be calibrated by a microbalance service technician
 according to the manufacturer's directions. The analyst should not attempt to repair the
 microbalance.

 The following general guidelines regarding set-up and placement of the microbalance should be
 followed to control environmental factors that may affect its performance.

   •   The microbalance should be stationary and level. If this does not occur, the microbalance
       will display instability and yield biased results. Therefore, it is imperative that the
       microbalance be located on a sturdy, vibration-free base (such as a stone weighing table)
       and away from any other equipment that  produces vibrations. If these arrangements are
       not possible, isolate the microbalance from such equipment by placing a stabilizing slab
       under the microbalance and/or by placing composite  vibration damping pads under the
       microbalance's legs or under the stabilizing slab.

   •   Ensure that the microbalance's base is sufficiently even to permit leveling of the
       microbalance according to the manufacturer's instructions. Most microbalances will have
       a bubble level that should be viewed to ensure that the microbalance is indeed level once
       it is set-up.

   •   It is essential that the microbalance be located away from sources of drafts, in order to
       ensure balance stability and proper readout. Therefore,  the microbalance should be
       located away from doors, windows, aisles with frequent traffic, ventilation ducts, and/or
       equipment with fans or moving parts. For example, a microbalance should not be placed
       in the flow path of a window air conditioning unit.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                   	January 2016    Page 4 of 14

   •   Locate the microbalance out of direct sunlight and away from local heating or cooling
       sources such as open flames, hot plates, water baths, ventilation ducts, windows, and
       heat-producing lamps.

   •   It is important for the microbalance to be properly grounded. A lack of appropriate
       grounding can cause significant stability issues with the balance. All microbalances have
       a grounding screw on the back that must be attached to a ground wire. See Figure 9.1 for
       an example. Additional information about static neutralization is found in Section 9.6.
                                                   snrtorius
                    Figure 9.1 Grounding screw on microbalance

 9.4   Logging Systems

 The weighing room's relative humidity (RH) and temperature conditions should be
 electronically measured and recorded on a continuous basis during filter conditioning, with 5-
 minutes values (at a minimum) logged and saved. There are a variety of RH and temperature
 sensors available on the market which come with electronic data-logging ability. Electronic data
 logging ability provides the user with simple and efficient options for manipulating data. Ensure
 that the sensors purchased meet the accuracy and resolution specifications detailed in Section
 4.3.8. The sensors should be recertified on an annual basis by the vendor to ensure accuracy and
 NIST-traceability.

 If the data recording device that communicates with the RH and temperature sensors requires
 selection of an interval upon initial set-up, the monitoring organization should select a 5-minute
 interval/average, at a minimum. More frequent measurements of temperature and humidity are
 encouraged. The collection of minute data increases the size of the data set available to be
 statistically analyzed, and will paint a better picture regarding the stability of the weighing
 room's environmental conditions. Lab analysts or managers can make more informed decisions
 regarding Appendix L compliance (see Section 10.4 of this document) when there is more data
 available to review.

 Some loggers record instantaneous readings of the temperature and relative humidity
 conditions, whereas others display an average value for the defined interval (e.g., 60 seconds
 averaged and reported as 1-minute data). The monitoring organization should define in its
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up
                                     January 2016    Page 5 of 14
 QAPP or SOP how its data recording device is programmed. Either scenario is acceptable,
 provided that the minimum amount of minute data is retained. Please note that raw data values,
 as opposed to rolling averages, should be used when computing the filter conditioning
 environment's 24-hour statistics (see Section 4.3.8).

 Many sensors will have accompanying software packages that will calculate the 24-hour
 averages and standard deviations from the collected data, making it easier for the lab analyst to
 determine compliance (or not) with the filter conditioning specifications. The software packages
 will often display those calculated results in tables, time-series graphs, or a combination of the
 two. See Figures 9.2 and 9.3 for two examples. If the sensors chosen do not have associated
 software that will calculate these statistics, the analyst will need to determine an efficient
 method to manually compute these  statistics.
               l HMTiJ?j
   *  ,•:. •
    PlM   TO  |21 OCTKiU  gj|

  31 QCT 30(4 • (SfS to Z2 OCT ISM • 0734
  MM   3S70
  Ml|r    *?9  CUITB* Riidmg
  M*    863  Migraud  X.SB
        (HT  \mmttmn:  38.43
   M*tn R*IH!M Humid*? l»
            )

             Kffl
             BO)
             3B(B
2»7-»
   PSI3
\maa»
 Figure 9.2: Example of environmental logging software displaying both summary
 statistics and a time-series graph
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up
January 2016     Page 6 of 14
                     Downloaded Data - Friday, November 14, 2014
           (TP125 )-THmp/LC Min: 21.B Mai: 22.B
                                              (TP125 )-RHra, Mm: 34.2 Max: 37.9
  37.5
  32.5
  27.5
         3 Sat
                                    11Tue
                                    DatefFime
                                             12 Wed
                                                      13Thu
                                                               14Fri
 Figure 9.3 Second example of environmental logging software display

 9.5  Data Management System

 An effective data management system is crucial in ensuring success in the laboratory and that
 data generated in the laboratory can be defended. The process for PM2.5 monitoring is data
 intensive and requires documenting and organizing a large amount of data to provide defensible
 evidence that required elements of the 40 CFR Part 50, Appendix L method have been followed
 and met.  The most efficient method of recording, storing, and manipulating this data is to use an
 electronic data management system.

 There are several commercially available software applications that are designed specifically for
 particulate matter weighing and are customizable for specific use cases. "In-house" designed
 databases may be used, but the developer must have detailed knowledge of the method and
 guidance to properly design and implement the system. A poorly designed data management
 system can omit  critical elements of the method that could result in data that is difficult to
 defend or deemed invalid because of missing components. Spreadsheets and paper copies are
 acceptable; however, these are extremely labor intensive in practical use when validating data
 and are not recommended. The QC requirements (RH differences, blank differences, check
 weight differences, etc) for PM2.5 filter weighing are very cumbersome to track without the help
 of an electronic system. These QC check results should be in a format that is easily reviewed
 during the weighing session to immediately assess the data quality.

 When selecting an electronic data management system (DMS),  there are several things that
 should be considered. The DMS should:
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                      January 2016     Page 7 of 14

    •   Address all critical Appendix L requirements;
    •   Address PM2.5 guidance, as applicable;
    •   Integrate room condition data (temperature and humidity logger);
    •   Include calibrations and reminders for weighing equipment (balances, loggers, weight
        standards);
    •   Provide QC check results in real-time during the weighing session;
    •   Aggregate and graphically display QC data (temperature, humidity, blanks, pre- and
        post-humidity difference, balance checks, etc); and,
    •   Use conditional checks/formatting to alert the analyst of potential issues.

 To support the data management system, the weigh room should be equipped with a computer
 that utilizes an LCD monitor. Older CRT monitors act as a reservoir for static; therefore, their
 use should be discouraged.  See Section 9.6 regarding the importance of static neutralization.

 9.6  Static Neutralization

 Static charge is the accumulation of electrical charges on the surface of a nonconductive
 material. In a PIVh.s gravimetric laboratory, electrostatic charge buildup will prevent a
 microbalance from operating properly. Common  symptoms of this problem include noisy
 readout, drift, and sudden readout shifts.  To reduce the risk of such interferences, efforts should
 be made to minimize electrostatic charge build-up within the filter conditioning area.

 Foremost, the microbalance itself should be properly grounded (see Section 9.3). The grounding
 screw on the back of the microbalance should be  attached to a grounding wire. It is also
 important  to ensure that the electrical outlet used  with the microbalance is properly grounded,
 too. The use of an electrical receptacle wall  plug AC outlet ground tester is recommended to
 check the  circuit. Please be aware, if the room chosen to serve as the filter conditioning
 environment is itself not properly grounded, it will cause instability issues with the
 microbalance that may be hard to diagnose and resolve when first detected. The AC  outlet
 ground tester is an easy, inexpensive way to double check the AC outlets in the selected
 weighing room.

 Do not assume that grounding eliminates all electrostatic buildup because the electrical ground
 may not be perfect. Even though a filter weight might stabilize within 30 to 60 seconds and no
 weight drift be observed during that period,  the microbalance may still be influenced by some
 electrostatic buildup.

 In addition to grounding the balance, other static control devices can be utilized to further
 reduce static charge. For example, polonium (210Po) is a radioactive material that is very
 effective in eliminating static. Radioactive anti-static strips containing a very small amount (i.e.,
 500 picocuries) of 210Po are safe and commonly available. 210Po anti-static strips will neutralize
 electrostatic charges on items brought within an inch of them. To reduce electrostatic buildup
 within the microbalance's weighing chamber, 210Po anti-static strip(s) can be affixed directly
 inside the  weighing chamber. Gooseneck holders containing 210Po anti-static strips can also be
 placed near the microbalance and used to neutralize individual filters prior to placement on the
 microbalance weigh pan. See Figure 9.4 for an example of these 210Po gooseneck holders.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up
                               January 2016     Page 8 of 14

210Po has a half-life of 138 days. If utilizing 210Po anti-static strips in the weighing room, they
will need to be replaced every 6 months. Dispose the old strips according to the manufacturer's
recommendations.

High voltage ionizing units use high voltage/low current charge to effectively neutralize
electrostatic charge on filters. See Figure 9.5 for an example of a U-electrode (ionizer) used in a
gravimetric laboratory. These units are safe and effective. They require periodic maintenance,
but unlike 210Po anti-static strips, they do not have to be replaced every 6 months. Similar to
using the 210Po anti-static strips gooseneck holders, an analyst will pass individual filters through
                                                             the ionizer bars of the U-
                                                             electrode in order to discharge
                                                             static prior to placement on the
                                                             microbalance weigh pan.

                                                             Other measures can be taken to
                                                             reduce sources of static in the
                                                             weighing room. For example,
                                                             anti-static solutions are
                                                             available for coating (and, at
                                                             appropriate and relatively
                                                             infrequent intervals, recoating)
                                                             the interior and exterior
                                                             nonmetallic surfaces of the
                                                             weighing room. This coating
                                                             facilitates the draining of
                                                             electrostatic charges away
                                                             from these surfaces (by
 making them conductive) to a common ground to which the metallic conductive surfaces are
 connected. Earth-grounded conductive mats may also be placed on the weighing table surface
 and beneath the analyst's shoe surfaces to reduce electrostatic charge  buildup. Grounding wires
 also can be attached to equilibration trays in order to dissipate static during filter conditioning.

 The lab analyst can use an anti-static spray to clean laboratory surfaces adjacent to the balance.
 The use of an anti-static brush should be employed to dust the microbalance weigh pan. The
 analyst should also be cognizant of any static-producing item(s) that might be on his/her person.
 For example, the analyst should avoid wearing wool or synthetic clothing that would hold a
 static charge. It is recommended that the analyst don an anti-static laboratory coat during filter
 weighing activities. If needed,  the analyst can wear a grounding strap  while weighing filters as
 well.

Charge  neutralization times may need to be longer than 60 seconds for sampling situations  in
which (1) a high amount of charge has developed on collected particles due  to their origin or (2)
the particle loading on a filter is large. Electrostatic charge buildup becomes greater as the air
becomes drier. A 60-second charge neutralization may be sufficient in  ambient indoor air
conditioned to 37 percent RH and 23 °C, but not 20 percent RH and 23 °C in arid environments.
This latter environment may require that the filter sit for more time on  the anti-static strip. The
longer neutralization period may have to be performed inside the weighing chamber or in a
 Figure 9.4: 210Po
 Anti-static strips
 & gooseneck Holders
Figure 9.5: U-Electrode
(ionizer)
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12. 9.0  Gravimetric Lab Design and Set-up                     January 2016    Page 9 of 14

second small chamber used only for charge neutralization. See Engelbrecht et al. (1980), Hawley
and Williams (1974), and Weil (1991) for more information about electrostatic charge and how
to minimize its effects.

It is important to reinforce that utilizing only one static control method may not be sufficient to
control static in all situations, and that a combination of these methods may be required. The
analyst should have a working understanding of static sources and controls that are specific to
the laboratory  and document the control measures accordingly.
 9.7  Mass Reference Standards

 Two separate sets of mass reference standards are recommended for use in the gravimetric
 laboratory: 1) primary standards, and 2) working standards. Procurement specifications for the
 mass reference standards are detailed in Section 4.3.7. Primary standards should be used for
 quarterly verification checks of the working standards. The working standards, on the other
 hand, are used for routine filter weighing and balance QC checks.

 Laboratory primary standards should be kept in the weighing room in a secure container which
 is clearly labeled as the "primary" weight set. Ideally, the primary standards should be stored in
 a separate location from the working standards (such as in a locked cabinet). Working standards
 should also be clearly labeled as the "working" weight set. The working standards should be
 kept next to the microbalance in a protective container.

 Both of these weight sets should be handled with care, in order to avoid damage that may alter
 their masses. To that end, the analyst should always use smooth, non-metallic forceps when
 handling the mass reference standards. The forceps used with the mass reference standards
 should not be used for any other purpose. Mark these forceps to distinguish them from the
 forceps used to handle filters. The forceps should be cleaned with alcohol and lint-free wipes
 and allowed to completely air dry before handling the mass reference standards.

 9.7.1   Certification Procedures - Primary Standards

 The laboratory primary mass reference standards should be certified every 365 days (annually)
 to a NIST-traceable reference standard of higher authority. The certification should occur at a
 State weight and measures laboratory or at another laboratory that is accredited by the National
 Voluntary Laboratory Accreditation Program (NVLAP), which is administered by NIST (Harris
 1994; White 1997). Certificates of traceability for the primary standards should be maintained
 in a central file within the monitoring organization, along with the certificates for other
 standards utilized within the laboratory. It is recommended that the analyst maintain copies of
 the certificates of traceability in the laboratory for quick access. Figure 9.6 provides a generic
 example of the information that may be presented on a certificate of traceability.

     NOTE:  Information and terminology  on calibration/traceability certificates may
     vary.  It is important that the lab analyst closely review the certificate provided by the
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.9.0  Gravimetric Lab Design and Set-up                     January 2016    Page 10 of 14

     vendor and understand the information represented. Most certificates will include
     definitions of key terms; however, if not supplied, such information may be requested
     from the vendor.

In Figure 9.6, the mass reference standard being recertified has a nominal value of 320 mg
(meaning, simply, the mass reference standard is supposed to weigh -320 mg).  The weight
certificate shows both "as found" and "as left" data (under the  "Conventional Mass Correction"
header). The nominal weight plus any correction factor becomes the conventional mass (i.e.,
certified weight) of the mass reference standard. In Figure 9.6, the "as left" correction factor for
the weight is 0.0102 mg. Therefore, the newly certified weight of this example mass reference
standard is, when rounded, 320.010 mg. The analyst should use the conventional mass (as
opposed to the nominal weight) as the certified (reference) value when utilizing this working
standard for routine QC checks.

Tolerance (also referred to as maximum permissible error) is the sum of the conventional mass
correction factor plus the measurement uncertainty of the metrology calibration laboratory's
ability to quantify that conventional mass. A weight standard would be defined as an ASTM
Class 0 check weight if the check weight had a tolerance of no more than 0.005 mg. In order for
the mass reference standard in Figure 9.6 to be defined as  an ASTM Class 0 standard, the sum of
the nominal weight, conventional mass correction factor, and the measurement uncertainty must
be 320 mg ± 0.005 mg (between 319.995 mg and 320.005 mg). Similarly, in order for the weight
standard to be defined as an ASTM Class 1  standard, the sum must be 320 mg ± 0.010 mg
(between 319.990 mg and 320.010 mg). For Class 2, the sum must be 320 mg ± 0.025 mg
(between 319.975 mg and 320.025 mg). Using the information supplied on the example
certificate (Figure 9.6), the summation of 320 mg (nominal value) + 0.0102mg (correction
factor) + 0.0018 mg (uncertainty) equals 320.012 mg. Therefore, the weight standard in this
example is classified as a Class 2 standard. The mass reference standards utilized in the
gravimetric laboratory should be Class 2 or better; standards with individual tolerances greater
than 0.025 mg should not be utilized in the PM2.5 laboratory.

The analyst should be cognizant of the "as found" versus "as left" status of the  weight standard,
as documented on the traceability certificate. If the mass reference  standard weighs significantly
less "as  found," it could indicate damage has occurred such that a portion of the weight standard
has chipped off (which could result from poor handling techniques by the analyst or from
damage sustained during transit/shipping to/from the metrology lab). If the mass reference
standard weighs more "as found," it could indicate contamination of the standard, or possibly
that the weight was incorrectly classified (i.e., Class 0, 1, or 2) during its previous certification.
In Figure 9.6, the traceability certificate indicates that the weight standard had to be adjusted
(corrected) because the "as found" conventional mass correction factor was significantly greater
than the tolerance allowed for Class 2 standards (0.025 mg). To "adjust" the mass of the weight
standard in this example, the metrology lab  may have cleaned  the weight standard, or possibly
shaved off a small portion of the weight. If the latter, the metrology lab would need permission
from the monitoring organization in order to make the necessary correction.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.9.0  Gravimetric Lab Design and Set-up                     January 2016     Page 11 of 14
      TRACEABLE CALIBRATION CERTIFICATE
                                                ANSI/NCSL Z540-M994; Pan 1 & ISO/IEC 17025 Accredited

      Client: ABC Monitoring Agency
      Date Received:                     12 FEB 2015
      Data Calibrated:                    18 FEB 2015
      Temperature Range:                  21.92 'C
      Prt»«ur« Rang*;                    734,2 nimHg
      Relative HumUHy Rang®:               55 %
      Air Density:                       , U99 ^n,.
      Tratsattte Report I;                   e2255525A
      NISI Certificate *:                    6811280358-10,6WI2844S1-14
      Tested By:                        08
      Procedure:                       Mass DiMemmallon {reSofsnet Hfl 9S2)
Nominal
Value Id,
ConvMitortil tiin..Con.
As Found As Left
(mg) (mi)
Une.1
K=2
(mg)
Tel.
(mo)
Balance
Usort
Standard Set Uscii
CattbratedWue
MM-DO-YY/MM-OD-YY
Assumed
Dfii5ity
(glcm')
         320 mg         1.2270     0.0102    IS 0.0018   0.025   650Q         01-22-15104-22-16
      Joe Metrobgist, February 20, 2015

 Figure 9.6:  Example calibration certificate for mass reference standard

 9.7.2  Verification Procedures - Working Standards

 Although it is suggested that the working standards be re-certified annually against a NIST-
 traceable standard at an accredited metrology laboratory, minimally, the working standards'
 masses should be verified against the laboratory's in-house primary standards every 90 days to
 check for mass shifts associated with handling or contamination. The verified values of the
 working standards as measured relative to the laboratory primary standards should be recorded
 in a laboratory QC notebook and used to check the calibration of the microbalance. If multiple
 microbalances are being used, all working standards should be verified at the same time to
 ensure that all gravimetric measurements are intercomparable.

The double substitution method is the recommended method for conducting quarterly
verifications of the working mass reference standards. This procedure is a version of SOP
Number 4 in NIST Handbook Number 14521. In this method, the laboratory analyst weighs a set
of primary standards against a set of working standards to generate a reference point. The
working and primary  standards are each weighed twice.

Whenever the double substitution is computed, the new calculation is compared to the previous
calculation to  determine if there has been a significant shift in mass. The double substitution
method is not  used to generate a "new mass" for any weight standard; it is intended only as a
verification (check) of the standards.
           j!i^^
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.9.0  Gravimetric Lab Design and Set-up                    January 2016     Page 12 of 14

The following is the recommended procedure. The analyst should weigh the standards at
regularly spaced time intervals to average out any effects of instrument drift.

1.  Zero and calibrate the microbalance following the microbalance's user guide. Exercise the
balance.

2.  Open the draft shield.  Using cleaned, non-metallic forceps, place the first working standard
(for example, a 400 mg weight), w, on the weigh pan.

3. Close the draft shield.   Wait until the display on the balance has stabilized.  Record the
weight as Measurement 1.

4. Open the draft shield and remove the weight.  Shut the draft shield and allow the
microbalance to zero.  Tare, if needed.

5.  Repeat steps 2-4 for the primary standard of the same weight (for example, 400 mg),/>, and
then the working standard, w, again to weigh each standard two times, recording the resulting
values as Measurements 2, 3, and 4, respectively.  Note that the primary standard is
consecutively weighed.
Measurement Number
1
2
3
4
Weight on Pan
w
P
P
w
Observation Number
0,
O2
03
04
  NOTE: The time intervals between successive trials should not differ from one another by
  more than 20%.  If this difference is exceeded, reject the data and take a new series of
  measurements.

6. Repeat steps 2-4 for the second mass reference standard utilized (for example, a 300 mg
weight).

7.  Calculate the mass correction, Cw, for the test (working standard weight, w) as follows,
according to the sequence used. In each case, the apparent mass correction for the primary
weight standard, Cp, are included. The symbols Np and Nw refer to the nominal values of p and
w, respectively.

                    CW = CP + ([Oi-o2+O4-o3]/2) + NP-NW

8.  Subsequent measurements of Cw must be within ±2  u.g of the initial Cw  value.

The analyst should document the results of this procedure in the laboratory logbook and/or on
any required data forms.
                                              2.12-PM2.5 Reference/Class I Equivalent Methods

-------
 2.12. 9.0  Gravimetric Lab Design and Set-up
                                                       January 2016      Page 13 of 14
Table 9-1.   Filter Preparation and Analysis Checks22
    Activity

 Microbalance
 Microbalance
 environment
 Mass
 reference
 standards
  Method and frequency
Working standards verified
every 3 months against
National Institute of Standards
and Technology (NIST)-
traceable laboratory primary
standards.
 Filter handling    Observe handling procedure.
 Filter integrity
 check
 Filter
 identification
 Pre- sampling
 filter
 conditioning
Visually inspect each filter.
Write filter number on filter
handling container and on
laboratory data form in
permanent ink.

Determine the correct
conditioning period  (at least
24 hours) for each new lot of
filters. Observe and record the
filter conditioning
environment's RH and
temperature on laboratory
data form.
     Requirements

Resolution of 1 (ig,
repeatability of 1 jig.

Climate-controlled,  draft-
free room or chamber or
equivalent; clean area
around microbalance.

Standards bracket weight of
filter; individual standard's
tolerance less than  25 |ig;
quarterly certification check
within ±2 ug  from Cw;
handle with smooth,
nonmetallic,  clean forceps.

Use powder-free gloves and
smooth, clean forceps.
Replace 210Po anti-static
strips every 6 months.

No pinholes, separation,
chaff, loose material, dis-
coloration, or filter non-
uniformity.

Make sure the numbers are
written legibly.
Perform lot blank stability
tests. Check for stability of
lot blank filter weights.
Weight changes should be
<15 (ig.  Mean RH between
30 and 40 percent, with a
variability of not more than
±5 percent over 24 hours.
Mean temperature should
be held  between 20 and
23 °C, with a variability of
not more than ±2 °C over 24
hours.
   Action if the require-
    ments are  not met
Obtain proper microbalance.
                                                          Modify the environment.
Obtain proper standards or
forceps.
                                                          Discard mishandled filter or
                                                          old anti-static strip.
Discard defective filter.
Replace label or correct form.
Revise conditioning period;
repeat conditioning.
 22 See Validation Template on AMTIC for any revisions to acceptance criteria in this table
 http://www3.cpa.gov/ttn/amtic/qapolli.itant.html
                                                                                              (continued)
                                                        2.12-PM25 Reference/Class I  Equivalent Methods

-------
 2.12. 9.0  Gravimetric Lab Design and Set-up
                                                  January 2016     Page 14 of 14
Table 9-1. (continued)
       Activity

    Pre- and post-
    sampling filter
    weighing
  Method and frequency

Observe all weighing
procedures. Perform all quality
control (QC) checks.
     Requirements

Neutralize the electrostatic
charge on filters. Wait until
balance indicates a stable
reading to record value.
Action if the require-
 ments are not met

Repeat weighing. Many
laboratories routinely
repeat the weighing.
    Internal QC        After every 10th filter, re-zero
                      the microbalance and reweigh
                      at least one working standard.
                      Weigh 10% laboratory
                      blanks/weighing session.
                      Reweigh one replicate filter at
                      the end of the weighing
                      session. Reweigh two working
                      standards at end of session.
                      Weigh 10% field blanks.
                               The working standard
                               measurements should
                               agree within 3 (ig of the
                               verified values. Laboratory
                               blank and replicate
                               measurements should
                               agree within 15 (ig.  Field
                               blank measurements should
                               agree within 30 jig.
                            Troubleshoot and take
                            appropriate corrective
                            action as specified in the
                            quality assurance
                            project plan (QAPP) to
                            attain acceptable levels.
                            Do not correct PM2.5
                            measurements based on
                            high blank levels.
    Post-sampling
    filter storage
Monitor the time between
sampling and weighing.
Weighing should be
completed within 240 hours
(10 days) after the end of
sampling, unless the filter is
maintained below average
ambient temp during the
entire time between retrieval
from the sampler and start
of the conditioning, in which
case the period shall not
exceed 30 days.
Troubleshoot and take
appropriate corrective
action.
    Post-sampling
    inspection,
    documentation,
    and verification
Examine the filter and field data
sheet for correct and complete
entries. If sample was shipped
in a cooled container, verify
that low temperature was
maintained.
No damage to filter. Field
data sheet complete.
Sampler worked OK.
Save filter for inspection.
Flag sample.
Troubleshoot and take
appropriate corrective
action.
    Post-sampling
    filter equilibration
Equilibrate filters for at least 24
hours. Observe and record the
filter conditioning environment's
RH and temperature on
laboratory data sheet.
Mean RH between 30 and
40 percent, with a variability
of not more than ±5 percent
over 24 hours, and within
5% RH of 24-hour average
pre-sampling RH. Mean
temperature should be held
between 20 and 23 °C, with
a variability of not more than
±2 °C over 24 hours.
Repeat equilibration.
                                                       2.12-PM25 Reference/Class I  Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January2016    Page1of23


 10.0 Filter Preparation  and Analysis

 10.1 Overview

 As discussed in Section 9.0, the quality of data from the PIVb.s sampling program depends on
 multiple factors. A crucial factor is the analytical laboratory staffs attention to detail and their
 microbalance operation technique. This section offers guidelines to enhance the data quality of
 the laboratory operation and, hence, the PIVb.s mass concentration and any additional chemical
 qualitative and quantitative determinations.

 Filter media requirements are given in Section 4.3.1.

 10.2   Filter Handling

Care in all aspects of filter/cassette handling cannot be overemphasized. The filters used for the
PM2.s program typically weigh between 150 - 400 mg (depending on filter manufacturer). Due to
the size and weight of the particles that will be collected on these filters, net weights will be
measured in micrograms (|ig). The loads on the filter may be anywhere from 10 to 2000 jig (83
|ig/m3) with most sample loads < 300 jig. In order to give one a sense of this weight, a 4 cm-long
human hair weighs -312 jig. A 300 jig sample load value represents -0.1- 0.2% of the weight of
the blank filter. In addition, it is expected that the laboratory analyst will be able to duplicate
weighings of the same filter to within 15 jig. A single thumbprint on a filter weighs  15 jig  or
more. It should be apparent that any small loss or gain (i.e., finger oils, dust) will affect filter
weights. The average volume  of a  24-hour sample is 24 m3 (16.67 L/min *1440
minutes). Therefore, a 300 jig net sample load would equate to a  concentration of 12.5 |ig/m,3
which exceeds the current  12.0 |ig/m3 annual NAAQS. Thirty micrograms of contamination
equates to approximately 1 |ig/m3  which, if it's a consistent source of contamination, can make a
difference in an attainment decision if sample concentrations are  near the NAAQS.

 Careful handling of the filter during sampling,  conditioning, and weighing is necessary to avoid
 measurement errors due to damaged filters or a gain or loss of collected particles on the filters.
 The analyst should wear clean laboratory clothing (a lab coat, e.g.) when working in the filter
 conditioning environment. Additionally, whenever filters are handled, the analyst should wear
 anti-static, powder-free gloves. These gloves act as an effective contamination barrier. Gloves
 that are packed in a box can carry an electrostatic charge. It is a good practice to discharge them
 by touching a good electrical ground after putting them on. The filters should be handled
 carefully by the support ring, rather than by the filter material, with smooth, non-serrated
 forceps that are used only for that purpose. Mark these forceps to distinguish them from the
 forceps used to handle mass reference standards. Forceps should be cleaned with alcohol  and
 lint-free wipes before handling filters and allowed to  air-dry. These precautions reduce the
 potential effect from body moisture or oils contacting the filters  and subsequently affecting the
 measured weights.

 In the laboratory, each filter should be transferred from its  sealed manufacturer's packaging to a
 clean filter-handling container, such as a glass or plastic petri dish or petri slide, to  reduce the
 risk of contamination.  The filter should remain in this container, except for weighing, until it is
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                         January2016     Page2of23

 loaded into a filter cassette prior to sampling. Each filter is required to have a unique
 identification number. Whether this identifier is a number or barcode, it is extremely important
 that filter suppliers use an identification system on filters and it is maintained throughout the
 sampling and analysis process in the monitoring networks. EPA's national filter procurement
 requires a unique filter identifier. It is recommended that each microbalance be assigned a block
 of filter numbers to be processed and used sequentially. Take care to avoid mistakenly assigning
 the same number twice or omitting a number. Color-coded labels may be helpful for separating
 groups of filters when more than one microbalance is being used.

 If sufficient resources are available, bar coding for filters can be introduced. Bar code readers
 and printers for microbalances are already generally  available and can significantly improve the
 efficiency of filter inventory tracking and processing.
 10.3   Filter Integrity Check

 EPA supplies filters manufactured from a single lot to the national PIVfo.s network. This lot of
 filters is performance tested prior to distribution to the monitoring organizations. However, all
 filters must be visually inspected at each weighing laboratory for defects before the initial
 weighing. If any defects are found, discard the filter. Specific filter defects to look for include
 the following:

  •    Pinhole—A small hole appearing (a) as a distinct and obvious bright point of light when
       examined over a light table or screen or (b) as a dark spot when viewed over a dark
       surface.

  •    Separation of ring—Any separation or lack of seal between the filter and the filter border
       reinforcing the ring.

  •    Chaff or flashing—Any extra material on the reinforcing, polyolefm ring or on the heat
       seal area that would prevent an airtight seal during sampling.

  •    Loose material—Any extra loose material or dirt particles on the filter.

  •    Discoloration—Any obvious discoloration that might be evidence of contamination.

  •    Filter nonuniformity—Any obvious visible nonuniformity in the appearance of the filter
       when viewed over a light table or black surface that might indicate gradations in porosity
       or density across the face of the filter.

  •    Other—A filter with any imperfection not described above,  such as irregular surfaces or
       other results of poor workmanship.

 Filter defects are expected periodically. Report to the EPA Regional Office if a high number of
 defects (e.g., >10%) are discovered in the lot of filters received by the monitoring organization
 throughout the year. Because filter defects are possible, a best practice is to order more filters
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                         January2016     Page3of23

 than needed according to historical observations. If an unusually high number of defects are
 reported, OAQPS can send additional filters to make up for the shortfall.

 10.4   Filter Conditioning

 Filter conditioning is a key activity in the gravimetric laboratory. Filters must be conditioned
 immediately before both pre- and post-sampling weighings. Filters must condition in the
 weighing room for a minimum of 24 hours to allow them sufficient time to equilibrate to the
 room's specific temperature and humidity conditions, which are prescribed in 40 CFRPart 50,
 Appendix L. It is imperative that the weigh lab analyst condition the PM2.5 filters in strict
 accordance with these requirements. Not meeting these requirements will be cause for data
 invalidation.

 10.4.1   Preparing the Filters

 10.4.1.1      Conditioning Prior to Sampling

 Foremost, a new lot of filters should be placed in the conditioning environment immediately
 upon receipt. Typically, filters come packed together in large groups or in a container with
 separators.  This package is usually contained inside another clear, reclosable plastic package,
 which may, in turn, be placed inside a box used in shipping. Filter lot stability testing should
 occur on the newly acquired filters. See Section 10.5 for more  details regarding lot stability
 testing.  The new lot of filters should be stored in the filter conditioning environment until the
 pre-sampling (tare) weighing activities are completed.

 To condition an unexposed (clean) batch of filters for their pre-sampling (tare) weighing, they
 should be removed individually from their sealed packages, inspected, and placed in their own
 filter-handling containers (e.g., petri dish or petri slide).  Place the filters in their filter-handling
 containers on a covered rack, conditioning tray, or an open-sided cabinet which will  allow air
 circulation  over the filters, while reducing the chance that airborne material inside the weighing
 room will settle onto the filters. See Figure 10.1 for an example. Label, if possible, both the
 filter handling container's  lid and bottom half. During conditioning, the lid should be placed
 either beneath the open container or partially covering the container. Place the filters on the
 conditioning trays (or rack/shelf) in numerical order so the analyst can more easily weigh the
 filters (in numerical order) during the upcoming initial weigh session. The filters must be
 conditioned in this manner for a minimum of 24 hours (or longer, depending on the results of
 the lot stability testing described in Section 10.5). This process allows the filter weights to
 stabilize before use.

    NOTE: Care should be taken to avoid contaminating PM2.5 filters inside the weighing
    room with particulates  released by other filter media (e.g., quartz and  glass) that are also
    being conditioned in the same area. Laboratory blanks (see Section  10.5) should be used
    to check for potential cross-contamination from airborne particulates inside the weighing
    room. If there is evidence of such cross-contamination, corrective actions should be
    taken. One possible solution for a cross-contamination problem is maintaining separate
    conditioning environments for PM2.5 filters and other filter media.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
                                                            January 2016    Page 4 of 23
10.4.1.2
           Conditioning After Sampling
Before conditioning exposed filters (i.e., ambient samples), the analyst should first determine the
requisite filter holding time for the individual samples as they are returned from the field. Filter
holding times - and how to determine if the final weigh date should occur within 10 or 30 days
post-sampling - are discussed in detail in Section 10.7. If shipments containing exposed filters
can be weighed promptly after arrival to the laboratory, then the analyst should begin post-
sample weighing activities (see  Section 10.7). If, however, the analyst will not be able to
immediately process and weigh the filters, then the analyst should place the samples into cold
storage (i.e., 1-4° C in a refrigerator) until such time that the post-sampling conditioning and
weighing activities can occur (but within the calculated filter holding time).

 It is important to note that exposed filters should only be placed in the weighing room for the
 amount of time necessary for them to equilibrate prior to final weighing. Keeping the filters in
 the weighing  room for longer periods of time may cause loss of volatiles from the filters, as well
 as subject them to additional contamination.  The maximum equilibration time for exposed
                                                   filters is recommended to be 72 hours.
                                                   With that in mind, the lab analyst should
                                                   not place exposed filters into the
                                                   weighing room for equilibration and
                                                   allow them to remain on their
                                                   conditioning trays for numerous days or
                                                   weeks prior to the final weigh.

                                                   To condition the exposed filters (after
                                                   they have been handled and inspected in
                                                   accordance with the procedures
                                                   described in Section 10.7), return them
                                                   to their assigned filter handling
                                                   containers  (e.g., petri dishes or petri
                                                   slides) and place on a covered rack,
                                                   conditioning tray, or an open-sided
                                                   cabinet that will  allow proper air
                                                   circulation. The filter container lid
                                                   should be placed either beneath the open
                                                   container or partially covering the
                                                   container. Place the filters on the
                                                   conditioning trays in numerical order so
                                                   the analyst can more easily  weigh the
                                                   filters (in numerical order) during the
                                                   final weigh session. Allow the filters to
                                                   condition in this manner for a minimum
                                                   of 24 hours (required), but not be more
 than 72 hours (recommended).  Under dire circumstances, exposed filters set out for
 conditioning can be returned to cold storage if an unexpected issue arises that hinders the
 analyst from final weighing within the requisite time frames. In case of this rare event, the
 reasons for returning the filters to cold storage should be documented in the laboratory logbook.
Figure 10.1  Filters equilibrating on conditioning
trays
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January2016    Page5of23

 For both pre- and post-sampling filter conditioning, the analyst should document the date and
 time that the filters are placed into conditioning in a laboratory logbook, at a minimum. Ideally,
 the dates/times of the conditioning period for each individual filter should be captured in the
 laboratory data management system, in order to provide evidence that the filters were
 conditioned in accordance with the reference method requirements.
 10.4.1.3 Determining Compliance with Filter Conditioning Specifications

 Prior to any pre- or post-sampling weigh session, the lab analyst must verify that the weighing
 room's environmental conditions have met the specifications of the PM2.5 reference method. If
 the room has not met all of the specifications of the method, the weigh session must be
 postponed. The reference method for PM2.5 (40 CFR Part 50, Appendix L) requires the
 following filter conditioning climate control:

       Section 8.2.1 Mean temperature.  20-23° C;

       Section 8.2.2 Temperature control.  ± 2° C over 24 hours;

       Section 8.2.3 Mean humidity.  Generally,  30-40% RH; however, where it can be shown
       that the  mean ambient relative humidity  during  sampling is  less  than  30 percent,
       conditioning is permissible at a mean relative humidity within 5 relative humidity percent
       of the mean ambient relative humidity during sampling, but not less than 20 percent;

       Section 8.2.4 Humidity control.  ±5 percent over 24 hours.
 This means that, during the 24-hour filter conditioning period23, the average temperature in the
 weighing room must be held within the range of 20-23°C. The temperature variability within
 that stated range cannot exceed ±2°C over the 24-hour period.  Similarly, the average RH must
 be held between 30-40 RH percent in the weighing room during the conditioning period. The
 variability in the RH cannot be more than ±5 RH percent over the 24-hour period. When
 samples are taken in arid locations, where RH values are consistently low, conditioning the
 filters in a more humid laboratory environment could potentially add water vapor to the samples
 and affect their final weight measurements. Therefore, it is permissible to condition those filters
 at an RH within 5 RH percent of the average ambient RH during sampling, but not less than 20
 percent. The temperature and relatively humidity control limits specified in 40 CFR Part 50,
 Appendix L, should not be confused to mean that the stated ranges  can be extended by ±2°C or
 ±5 RH%, respectively. For example, if the 24-hour temperature average is found to be 18.5 °C,
 then the Appendix L requirement for temperature has not been met, because the  24-hour
 average must fall within the range of 20-23 °C. Similarly, if the 24-hour relative humidity
 average is found to be 42% RH, then the Appendix L requirement for relative humidity has not
 been met because the average must fall within the range of 30-40% RH.
23 A filter conditioning period may be longer than 24 hours. However, for accepting laboratory temperature and
humidity control, the last 24 hours prior to initiation of weighing is considered the filter conditioning period.


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
January 2016     Page 6 of 23
 The monitoring organization is afforded some flexibility in determining how to demonstrate the
 room's climate conditions have stayed in control over the 24-hour period. The monitoring
 organization must describe in its QAPP and SOP how it will define and demonstrate control
 within the gravimetric laboratory. The use of standard deviation (SD), using the 5-minute
 temperature and RH values as recommended in Section 9.4, is recommended as the best means
 to demonstrate the room has met the control requirements stipulated in the reference method.
 The standard deviation calculation will minimize the impact of spikes in relative humidity and
 temperature readings, and therefore provide an indicator as to the true variability in the
 weighing room during filter conditioning. If the  calculated standard deviation exceeds the limits
 stated in the reference method, then the weighing room has not demonstrated control and any
 pending weigh session(s) must be postponed.
so
45
40
35
30
25
20
15
10
5
0
Scenario #1 RH (Mean 37; SD 3.4)



















>. ^-~













































1 2 3 4 S 6 7 8 9

-




II




















—_^— ^






11 12 13 14




















15 16 17 18 19











20 21 22 23
                                                Even if the prior 24-hour temperature and
                                                humidity range and SD calculations meet
                                                acceptance criteria, spikes of temperature or
                                                RH may occur.  Figure 10.2 shows two
                                                examples of relative humidity over a 24-
                                                hour filter equilibration period using hourly
                                                averages. In both scenarios, the average 24-
                                                hour RH (37%) and the standard deviation
                                                (3.4%) meet the method requirements.
                                                However, in Scenario #1, there are a few
                                                hours at the beginning of the equilibration
                                                period where the RH is outside of the
                                                humidity specification. In Scenario #2, there
                                                are a few hours towards the end of the
                                                equilibration period where the RH is out of
                                                the humidity specification. In Scenario #1,
                                                the laboratory analyst may decide to accept
                                                the spikes at the beginning of the session
                                                and allow weighing to commence (because
                                                the 24-hour average and  SD are both within
                                                specification). However, in Scenario #2,
                                                even though the 24-hour average and SD
                                 _               are within  specifications, the laboratory
  Figure 10.2 Relative humidity (RH) scenarios   analygt may take a more conservative
 approach and not allow weighing to commence because the RH spikes occurred at the end of
 the equilibration period, which would reflect the current condition of the filters. When assessing
 temperature or relative humidity spikes, the appropriate local decision maker should evaluate all
 relevant data and decide if the spikes are significant enough to adversely affect filter weights.
 Experience with this should eventually lead to the ability to estimate which spikes are too big or
 too lengthy. The evaluation should include the significance of the timing of the spike in relation
 to the time of weighing, with the closer to the weighing, the more significant the effect on the
 filter weight.

 However, if the spikes in either scenario from Figure 10.2 had been more drastic - such that the
 calculated SD were greater than 5% RH - then the filter conditioning environment would not
Scenario #2 RH (Mean 37; SD 3.4)
50
45
35
30







"X-~v^_^^^^_^-—"















































































































^^








1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21







22 23
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January2016    Page7of23

 meet the Appendix L criterion for humidity control, regardless of the timing of the spikes during
 the 24-hour period. In this case, filters should remain in the conditioning environment until the
 24-hour environmental conditions are met before weighing commences. It is important to note
 that the statistics calculated for the 24-hour filter conditioning period should be computed from
 the data collected for the 24 hours prior to the start of the actual weighing session, instead of the
 midnight-to-midnight time frame prior to the start of the weigh session. For example, if the
 weighing session is slated to start at 9am on Monday, then the lab analyst should review the
 weighing room's environmental  statistics from 9am Sunday until 9am Monday and verify that
 the temperature and relative humidity conditions meet reference method specifications before
 proceeding with the weighing session.

 The statistics computed from the environmental sensors in the laboratory to determine
 compliance with the filter conditioning requirements must be recorded and saved, preferably in
 the laboratory's data management system. At a minimum, the lab analyst should document the
 summary statistics that represent the filter conditioning period's environmental conditions in a
 laboratory logbook, as well as the dates/times of the  conditioning period itself. If possible, it is
 suggested that the lab analyst directly download the minute data obtained from the
 environmental sensors and calculate the  24-hour averages and standard deviation manually.
 This process would serve as a means of verifying that the weighing room has met method
 specifications, as well as verifying the correctness of the calculations performed by the logging
 software. Microsoft Excel, or a similar spreadsheet, can be used to easily calculate means and
 standard deviations of the collected minute  data, as well as create control charts. It is
 recommended that the lab analyst save this  spreadsheet as further evidence that the filter
 conditioning environment met the specifications of Appendix L and the monitoring
 organization's QAPP.

 Along with saving and documenting filter conditioning temperature and RH measurements, any
 malfunctions, discrepancies, and/or maintenance activities that occurred in the weighing room
 during filter conditioning should be recorded in a logbook or  a laboratory notebook.

 Pre- and Post-Sampling Filter Conditioning Comparison

 In addition to determining if the  24-hour filter conditioning period prior to a weigh session
 meets Appendix L specifications, the reference method also calls for filters to be conditioned at
 the same relative humidity conditions during their pre- and post-sampling weighings.
 Specifically, Appendix L states:

      Section  8.3.3. Filters must  be conditioned at the same conditions (humidity within  ±5
      relative humidity percent) before both the pre- and post-sampling weighings.

 This additional relative humidity requirement is established in order to limit the effects of water
 vapor between pre- and post-sampling weighing sessions. To help clarify this requirement,
 consider the following two scenarios:

   A. During the pre-weigh filter conditioning of a batch of filters, the average RH is 33%.
      During the post-weigh conditioning of the same batch of filters, the average RH is 36%.
      The RH difference between these  two weighing sessions is 3% RH. Under this scenario,
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
January 2016    Page 8 of 23
       the pre- and post-sampling weighings meet the Appendix L requirement.

   B.  During the pre-weigh filter conditioning of a batch of filters, the average RH is 33%.
       During the post-weigh conditioning of the same batch of filters, the average RH is 39%.
       The RH difference between these two weighing sessions is 6% RH. Under this scenario,
       the pre- and post-sampling weighings do not meet the Appendix L requirement.

 In order to ensure that the pre- and post-weighing sessions are within ±5% RH, the
 environmental conditions in the filter conditioning environment must be tightly controlled and
 closely monitored. To this end, charting the temperature and relative humidity statistics in the
 weighing room is recommended. If the logging software utilized in the laboratory does not have
 this capability, then the lab analyst is encouraged to export the data collected from the logging
 system and enter it into a spreadsheet (or similar) and create control charts or time-series graphs
 that demonstrate trends in the laboratory environmental conditions. Figure 10.3 gives an
 example of daily 24-hour averages plotted on a time-series graph in order to assess trends in
 humidity over the course of a quarter. This graph shows swings in relatively humidity
 conditions, which may make it difficult for filter weighings to pass the pre- and post-RH
 requirement. Charts such as this will provide the lab analyst a quick tool to easily assess overall
 stability in the weighing room. Recurring spikes  in the weigh room's temperature or relative
 humidity conditions should be investigated, as they could indicate multiple issues (such as a
 faltering HVAC system) that would warrant corrective actions.
                               Daily Means
    £45
    S 40
           147 10131619222528313437404346495255586164677073767982858891
                                           DAYS
 Figure 10.3 Laboratory relative humidity conditions during a quarter
 Tracking the pre- and post-weigh RH conditions can be challenging for the lab analyst. To make
 this task easier, it is recommended that the databases or spreadsheets utilized in the laboratory
 be formatted to provide some type of warning indicator to alert the analyst that this specific RH
 criterion may be exceeded if a post-sampling weigh session commences.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January2016     Page9of23

 10.4.1.4      Sample Batching

 In the context of this document a sample batch is defined as a grouping of filters that are
 conditioned and analyzed as a single event for which laboratory quality control requirements
 (i.e., temperature, relative humidity, etc) apply. The batch includes the required QC filters
 (blanks, duplicates, etc). A batch may be based on the number of filters that can be processed in
 a normal business day while considering post-sampling holding times, as well as ensuring that
 the post-sampling 24-hour average RH is within lab requirements and is within +5% of pre-
 sampling RH.

 10.5   Quality Control Blanks

 There are four types of filter blanks that are commonly used in PM2.5 monitoring: field blanks,
 laboratory blanks, lot blanks, and trip blanks. Field and laboratory blanks are required by 40
 CFR Part 50, Appendix L; lot blanks and trip blanks are QC blanks that are not required, but are
 considered good laboratory practice for conducting assessments. Descriptions of the four types
 of blanks and their uses are as follows:

Field Blank - Collecting field blanks is required under 40 CFR Part 50, Appendix L, Section
8.3.7.1. A field blank is a filter that is pre-weighed with routine  samples, installed in the field
sampler without any flow passing over the filter, re-weighed with routine samples, and then
initial/final weights compared. The purpose of field blanks is to provide an estimate of total
measurement system contamination, including laboratory and field activities. Through a
comparison of laboratory blanks against field blanks, contamination from field activities can be
assessed. The validation acceptance criterion for field blanks is ±30 jig between weighings. The
pre- and post-sampling weights and weight difference should be recorded in the QC notebook,
on the laboratory data form, and/or in an equivalent database.  If the field blank weight change
exceeds 30 ug, implement appropriate troubleshooting and corrective actions.

In practice, field blanks for single channel samplers should be transported to the sampling site,
momentarily installed in the sampler, removed, and then stored in their protective containers
inside the sampler's case at the sampling site until the exposed filters are retrieved for post-
sampling weighing. In the case of sequential (i.e., multi-filter) samplers, field blanks should be
placed in unused sampler slots, kept there during the sampling interval, and retrieved with the
exposed filters. This approach presumes that unused slots are available and that the sampler can
be programmed to not sample the field blank. If this cannot be done, follow the procedure used
for the single channel sampler. As a recommended best-case practice, site operators are
encouraged to occasionally install field blanks in a non-active sampler's filter holder over a 24-
hour period during which the sampler is not scheduled to be sampling.  This practice may be a
useful troubleshooting technique.

Field blanks should be collected at a frequency of 10% of the  sampling runs scheduled per site.
For example, a monitor operating on a l-in-6 day schedule would be expected to have 6 blanks
in a year, while a monitor operating every day would be  expected to have 36 blanks. Field blanks
should be taken throughout the duration of the sampling schedule (spaced evenly across the year)
and not concentrated in a short period of time.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January 2016     Page 10 of 23

To best utilize field blanks, the measurements should be control charted over time to identify and
address trends before the acceptance criterion is exceeded. Field blank control charts should also
be periodically compared to laboratory blank control charts to determine contamination
originating from field activities. Since field blanks reflect the effect of factors occurring at all
steps of the data collection process, evaluation of the effects of field factors requires removing
the lab blank variability from the total measurement system variability, and then looking at the
remaining values. Field variability could be due to factors having an effect on the cassette
assembly, transportation to/from the field site, or issues with the sampler itself (such as sampler
cleanliness).

Laboratory Blank - Weighing laboratory blanks is required under 40 CFR Part 50, Appendix L,
Section 8.3.7.2. A laboratory blank is a filter that is inspected, equilibrated, pre-weighed with  a
specific batch of filters, post-weighed with the same batch or subset of that batch of filters, and
the two weights compared. The pre- and post-sampling weights and weight change should be
recorded in the QC notebook, on laboratory data forms (see Figure  10.5), and/or in an equivalent
database. The purpose of the laboratory blank is to mimic the laboratory conditions to which all
filters in the batch are exposed during inspection, pre-  and post-equilibration, and pre- and post-
weighing session activities. Laboratory blanks provide an estimate of contamination that may  be
occurring in the laboratory during these activities. The validation  acceptance criterion for
laboratory blanks is ±15 jig between weighings. If the  laboratory  blank weight change exceeds
15 ug, implement appropriate troubleshooting and corrective actions.

Laboratory blanks are intended to be single use. The laboratory analyst must plan to designate
several filters as laboratory blanks during the pre-weigh process because all  filters from the batch
may not all be weighed in the same post-weigh session. Therefore, enough laboratory blanks
should be weighed during a pre-sampling weighing session to provide at least one single-use
laboratory blank during each subsequent post-sampling weighing session.  Typically, assigning
10% of the total  filters in a batch as laboratory blanks will allow for enough single-use blanks.

At least one laboratory blank from a pre-weigh session must physically remain with the same
batch through all laboratory  activities. However, when that batch  of filters is in the field, the
laboratory blanks must be kept completely covered in the weighing room until the filters return
to the laboratory. The rationale for this procedure is, since the batch of filters is in the field and
not exposed to any potential laboratory contamination, the laboratory  blanks should be covered
to mimic the absence of the batch from the weighing laboratory. When the filter batch arrives
back to the laboratory, the laboratory blanks are returned to the batch, uncovered, equilibrated,
and weighed with the batch.

To best utilize laboratory blanks, the measurements should be control charted over time to
identify and address trends before the acceptance criterion is exceeded. See Figure 10.4 for an
example control  chart. Assessing elevated laboratory blanks requires examination of all
laboratory processes, equipment, and supplies including filter handling, laboratory cleanliness,
HVAC filter effectiveness, and shipping media cleanliness.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
January2016     Page11of23
                                   LAB BLANKS
   0030
   0025
  I 0.020
  S 0015
 I" 0010
 * 0.005
  « 0000
  | -° °05
 I -0.010
 0-0.015
   -0.020
   -0025
   -0030
        N JN R K S ty N
        i 1 s s i s i
                   2 §
                                    Final Weight Date
  Figure 10.4 Lab blank control chart


Lot Blanks - The use of lot blanks is not required by 40 CFR Part 50, Appendix L; however,
their use is considered a "best practice". A filter "lot" may be defined as a group of filters from
the same manufacturer, manufactured using the same materials and process, and having the same
physical and chemical characteristics. Different lots may have slightly different physical and
chemical attributes that may affect equilibration times, such as differences in Teflon® or support
ring thicknesses. A lot blank is a filter from the lot that is used to determine filter weight stability
over a period of time due to the volatilization of material from the filter, or the absorption of
gaseous material into the filter from the atmosphere. In general, lot blanks are weighed
repeatedly over time to show when stability is achieved.

The most  common use of lot blank filters  is in a filter lot stability test. Each year, EPA OAQPS
provides new acceptance-tested Teflon® filters from a single lot to agencies conducting PM2.5
monitoring for NAAQS. 40 CFR Part 50,  Appendix L, Section 8.2 requires all filters to be pre-
and post-conditioned (equilibrated) in accordance with specified temperature and relative
humidity control requirements. Section 8.2.5 of Appendix L sets the minimum conditioning time
at 24 hours.  However, every laboratory is different and has different environmental set-points
which could have an effect on filter equilibration. Lot stability testing is used to determine the
optimal time required to equilibrate the new lot of filters to the specific laboratory with its
specific environmental conditions. Resultantly, EPA recommends the use of lot blanks for this
particular purpose in the PM2.5 gravimetric laboratory.

To complete a lot stability test, nine filters from the new lot are weighed - three filters from
three boxes  selected at random - every 24 hours until the weight change of each filter is less than
15 ug (i.e., three times the precision for weighing unexposed filters).  This experiment determines
the period of time that the filter lot should be conditioned in the specific laboratory prior to
weighing. The results of this one test will  be used for all filters in the lot until the lot is depleted.
The experiment should be repeated whenever a new lot of filters is received.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January 2016    Page 12 of 23

   NOTE:  Researchers have found that the precision of mass measurements for unexposed
   filters based on replicate weighings is typically 3 to 5 jig/filter (Desert Research Institute
   1994). The precision on exposed filters is typically 5 to 8 jig/filter. Precision for exposed
   filters with loadings heavier than 1 mg/cm2 of filter surface area may approach ±2 percent
   of the loading.  The precision for net mass filter loadings (typically 6 to 9 jig/filter) is
   defined as the square root of the sum of the squares of the pre- and post-sampling
   precision.

In addition to the lot stability testing, lot blanks can be utilized in the gravimetric laboratory as a
tool for long-term contamination testing. A long-term exposure blank is a filter, or group of
filters, from the lot that is set out in the weighing room in a constant state of conditioning and re-
weighed periodically (e.g., weekly) to observe weight change over an extended period of time
(e.g.,  6 months to a year). If the blanks show a steady weight increase, corrective action may be
necessary to control contamination in the laboratory more effectively. However, the use of long-
term exposure blanks is optional. The monitoring organization should decide whether or not such
testing would be beneficial to its specific QA Program.
Trip Blank - Collecting trip blanks is not a requirement under 40 CFR Part 50, Appendix L;
however, collecting trips blanks is a "best practice". A trip blank is a filter that is treated exactly
as a field blank, but it is never placed into the sampler or exposed to the ambient environment.
The purpose of the trip blank is to assess possible contamination to filters during packing and
transport to and from the laboratory to the sampling location. The pre- and post-sampling
weights should be recorded in the QC notebook, on the laboratory data form (see Figure 10.5),
and/or in an equivalent database. The validation acceptance criterion for trip blanks is ±15 jig
between weighings. If the weight change exceeds 15 ug,  contamination in the laboratory or
during shipping may be occurring. Implement appropriate troubleshooting and corrective actions.

In practice, a trip blank travels with the filters that are used in sampling, but the trip blanks
remain protected at all times and are never left in the sampler. In some cases, this may mean that
the trip blank is protected in an anti-static bag or with a cap.  In other instances, where the filters
are shipped in a magazine, the trip blank may travel on top of the filter stack, be removed and
capped before installation, and then taken back to the field office. In every case, the trip blank is
always protected, is never installed in the sampler, and is never left in the sampler.

When utilized, trip blanks are typically collected at a frequency of 5% of the scheduled sampling
runs per site,  and are taken throughout the duration of the sampling schedule (evenly spaced
across the year). However, trip blanks can be used at a higher frequency or on a  short-term basis
(i.e., concentrated period of time) as an investigative tool when field and laboratory blanks are
elevated.

To best  utilize trip  blanks, the measurements should be control charted over time to identify and
address  trends before the acceptance criterion is exceeded. Trip blank control charts should also
be periodically compared to laboratory blank control charts to determine contamination
originating from packing and shipping.  Ideally, laboratory and trip blank data should be
identical, meaning  there is no impact from packing and shipping.  Residue from cassette cleaning
or particles in shipping bags can elevate field blanks and could be falsely identified as a field
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
January 2016     Page 13 of 23
handling or sampler cleanliness issue. Trip blanks compared to both field and laboratory blanks
could easily isolate the problem so corrective action could be taken.
Filter Lot ID:
Balance ID:
C20102
A44603
Date: 6-29-14
Time: 10:00 AM
Weigh Room Conditions
RH 24-hr Avg: 33.52 SD: 1.4
Temp 24-hr Avg: 20.74 SD: 0.2
Analyst: John Smith
SOOmg Certified %QQ QQ2
Weight:
SOOmg Certified 500.002
Weight:
Weigh Session Type: INITIAL WEIGH
Filter ID

300 mg
SOOmg
T4605642
T4605643
T4605644
T4605645
T4605646
T4605647
T4605648
T4605649
T4605650
T4605652
SOOmg
T4605653
T4605654
T4605655
T4605656
T4605657
T4605658
T4605659
T4605660
T4605661
T4605662
SOOmg
T4605663
T4605664
T4605601
T4605651
T4605642
SOOmg
SOOmg
Lab Analyst:
Verified by:
Filter Type*

R
R
F
F
F
F
F
F
F
F
F
F
R
F
F
F
F
F
F
F
F
F
F
R
F
F
Lab Blank
Lab Blank
Duplicate
R
R
Weight
300.003
500.003
375.031
375.498
375.913
383.603
383.249
375.56
376.275
371.318
371 .346
381.318
300.002
376.338
380.271
376.027
374.86
369.491
373.477
373.09
378.984
378.768
379.966
300.002
372.631
372.523
371 .856
374.324
375.031
299.999
500.003
John Smith
Joe Supervisor
QC/Check Weight
Difference (mg)
0.001
0.001

0.000

0.000

-0.003
0.001
Criteria
Passed?
Y
Y

Y

Y

Y
Y
Date: 6/29/2014
Date: 6/30/2014
*R= Mass Reference Standard; *F = Filter

 Figure 10.5 Example laboratory data form.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                       January 2016    Page 14 of 23

 10.6   Pre-Sampling Filter Weighing (Tare Weight)

 40 CFR Part 50, Appendix L, Section 8.3.5 requires that the pre-sample filter weighing be
 conducted within 30 days of the sampling period. Section 8.3.2 of Appendix L requires the
 analytical balance to be located in the same controlled environment in which the filters are
 conditioned; additionally, the filters are required to be weighed immediately following the
 conditioning period without intermediate or transient exposure to other conditions or
 environments.

 Filters should be weighed and processed in batches. See Section  10.4.1.4. Simply put, a batch is
 a set of filters awaiting pre- or post-weighing.  There is no magic number to the size of a batch,
 since it depends on the size of the laboratory, the type of equipment in use in the laboratory, and
 the size of the PM2.5 network(s) for which the laboratory processes filters. However, a batch of
 filters should be considered as the group of filters that can be weighed, following requisite
 conditioning, in a  single business day, since the 30-day pre-sampling filter holding time starts
 once a filter has been pre-weighed.

 NOTE:  One might consider a weighing event to include both the filter conditioning process
 (lasting approximately 24-72 hours) and the filter weighing session (which should occur during
 the span of a typical 8-10 hour business day). In this scenario, a weighing event could total
 approximately  2-4 business days. However, when thinking of the weighing session, the
 gravimetric analysis of a batch of filters should occur no later than 10 hours from the end of the
 conditioning period. In other words, a batch of filters should be weighed within the same
 business day that the filter conditioning period is documented as  ending (and for which 24-hour
 environmental  statistics have been computed).  This guidance distinguishes the conditioning
 period from the weighing session for convenience of discussion only. It expected that during a
 weighing session the laboratory temperature and humidity conditions are maintained within the
 required specifications of 40 CFR Part 50, Appendix L, Section 8.

 When preparing a batch of filters to be weighed, the analyst should include the necessary
 number of QC  filters - lab blanks and field blanks.  The analyst should plan to weigh enough
 laboratory blanks during a pre-sampling weighing session to provide at least one single-use
 laboratory blank during each subsequent post-sampling weighing session. Typically, assigning
 10% of the total filters in a batch as laboratory blanks will allow  for enough single-use blanks.
 As an example, if  a batch is prepared to include 100 routine sample filters, the analyst should
 designate -10 filters as lab blanks. The number of field blanks to prepare, however, depends
 largely on the sampling schedules of the samplers in the PM2.5 network. In some cases, the
 monitoring organization will request the specific number of field blanks needed. It is also good
 practice to include several extra filters in a batch to account for unforeseen circumstances such
 as filter mishandling in the lab or make-up sample runs.

 Once a batch of filters is defined and properly conditioned,  the pre-sampling (tare) weighing
 session can begin. The steps that follow present procedures specific to a common commercially
 available microbalance. Calibration, QC checks, and operational procedures may have to be
 adapted for specific microbalance models. If the monitoring organization utilizes a
 balance/autohandler system, the procedures stated herein may need to be adapted for that
 specific system as well. If using an autohandler, the laboratory analyst is still responsible for
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                       January 2016    Page 15 of 23

 ensuring that all of the filter conditioning and weighing requirements (including QC checks) are
 successfully met and documented, as specified in 40 CFR Part 50, Appendix L. Additionally,
 the laboratory analyst charged with overseeing the autohandler operation and maintenance
 should be fluent in manual weighing procedures, and be prepared to weigh filters manually (as
 described in Sections 10.6 and 10.8 of this document), in the event of autohandler downtime.

 To ensure maximum stability, it is recommended that the microbalance be turned on at all
 times. This enables the microbalance to be operational at any time and eliminates the need for a
 warm-up period before analyses are performed. Newer microbalances are always turned on
 (except for their displays) when they are plugged in, but may revert to a "sleep mode" during
 inactivity.

 Follow these steps during the pre-sampling weighing session:

 1.   Don an anti-static laboratory coat and anti-static, powder-free gloves.

 2.   Verify that the weighing room's environmental conditions are in compliance with
     Appendix L requirements (see Section 10.4). Record the 24-hour averages for RH and
     temperature on the laboratory data form and in the laboratory QC notebook or database.
     The SD of these parameters should also be documented.

 3.   Clean the microbalance's weighing  chamber with a fine brush. (Anti-static brushes for
     cleaning the microbalance and the surrounding area are available. Avoid using pressurized
     gas, which may blow damaging debris and oils into the microbalance's mechanism.)

 4.   Clean the surfaces near the microbalance with anti-static solution or alcohol-moistened
     disposable laboratory wipes.

 5.   Clean both the mass reference standards forceps and the filter forceps with a lint-free cloth
     and moistened laboratory wipes. Allow the forceps to thoroughly air-dry. Even a small
     amount of moisture can cause a significant measurement bias.

 6.   Zero (i.e., tare) and calibrate the microbalance according to the manufacturer's directions.
     Many newer microbalances calibrate themselves automatically when the operator presses a
     key.

7.   Exercise the microbalance by opening and closing the draft shield five times. This
     encourages smooth operation of the draft shield mechanics and allows air from the
     equilibrated environment to enter into the balance.

8.   Using smooth, non-serrated,  nonmetallic forceps, weigh two working mass reference
     standards - one 100 mg lighter and one  100 mg heavier than an average unloaded filter - as
     a QC check. Handle the working standards  carefully to avoid damage that may alter their
     masses. Wait until the microbalance's display indicates that a stable reading has been
     obtained. (Consult the microbalance's operating manual for more information about
     obtaining stable readings.) Record both the verified and measured values of these standards
     on the laboratory data form and/or in the laboratory QC notebook or database.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January 2016    Page 16 of 23
     If the verified and measured values of a working standard disagree by more than 3 ug (i.e.,
     three times the microbalance's repeatability), reweigh the working standard. If the two
     values still disagree, troubleshoot and take appropriate corrective action, which may
     include (1) reverifying the working standards against the laboratory primary standards
     and/or (2) having a  service technician repair the microbalance. The analyst should not
     attempt to repair the microbalance.

9.    Weigh the filters. Operate the balance according to the manufacturer's directions.

        a. Zero the microbalance. The preferred way to zero the microbalance is to allow the
        balance to return to zero on its own. The microbalance may be manually tared to force
        zero, but this practice may mask drift or stability issues with the microbalance.

        b.  Take the filter from its filter-handling container (petri dish or petri slide) by gently
        slipping the filter-handling forceps under the outer polyolefin support ring.

        c.  Hold the filter only by the ring, not by the filter material. Pass the filter near an anti-
        static device (see Section 9.6) for 30 to 60 seconds immediately prior to weighing. The
        anti-static strip should be inside the microbalance's weighing chamber or as close to the
        chamber door as  is practical.

        d.  Immediately  transfer the filter to the microbalance's weigh pan and close the
        weighing chamber door.

        e.  After the microbalance's display indicates that a stable reading has been obtained,
        record the balance number, the filter number, the filter lot number,  and the filter's tare
        weight (pre-sampling mass) on the laboratory data form and/or in the laboratory
        notebook or database.

        f  Return the filter to its filter-handling container and place the lid on the container.

10.  After every 10th filter, the analyst should reweigh at least one of the working standards.
     Record the measurement on the laboratory data form and/or in the laboratory QC notebook
     or database. If this measurement  disagrees from the verified value by more than 3  ug,
     reweigh the standard. If the two measurements still disagree, troubleshoot and take
     appropriate corrective action, which may include (1) reweighing all of the previously
     weighed filters, (2)  reverifying the working standards against the laboratory primary
     standards, and/or (3) having a service technician repair the microbalance. The analyst
     should not attempt to repair the microbalance.

11.  At the end of the weighing session, reweigh both working standards. Record the
     measurements on the laboratory data form and/or in the laboratory QC notebook or
     database.

12.  Weigh the batch duplicate filter as a final QC check. The batch duplicate is one routine
     filter randomly selected from the batch and reweighed; the acceptance criterion is ±15 ug.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                        January 2016    Page 17 of 23

     Record the duplicate measurement on the laboratory data form and/or in the laboratory QC
     notebook or database. If the batch duplicate filter reweigh is within acceptance criteria,
     move on to step 13. If the duplicate measurement disagrees from the original measurement
     by more than 15 ug, reweigh the filter. If the measurements still disagree, troubleshoot and
     take appropriate corrective action, which may include:

          a)  reweighing all or some of the previously weighed filters to determine if the initial
              filter chosen was an anomaly or the batch has an integrity issue,
          b)  reweighing the working standards and primary standards to ensure that balance is
              operating properly and, if not, having a service technician repair the microbalance
              (the analyst should not attempt to repair the microbalance), and
          c)  reviewing lab temperature and humidity conditions during the weighing event to
              determine if the batch was somehow effected. The monitoring organization's
              QAPP and SOPs may provide additional troubleshooting techniques based on
              experience in the laboratory.

 13.  Return the duplicate filter to its filter-handling container. Replace the lid to cover the filter.
     Return all filters in the batch,  with  lids closed, to their conditioning trays/shelves.

 14. The analyst should notify his/her supervisor that the weighing session is complete. The QC
     or laboratory supervisor should then review the results of the weigh session. The QC or
     laboratory  supervisor should verify on the laboratory data form and/or in the laboratory
     notebook or database the acceptability of the weighing session and QC checks, as well as
     the compliance with the weighing room's environmental control statistics. The weights  of
     the filters should be reviewed; any  routine filter whose weight is outside the normal range
     should be investigated and possibly removed from use. The QC or laboratory supervisor
     should sign or initial the data  form, notebook, or database entry, in order to document the
     weigh session has been verified and approved.

 As soon as possible after the pre-sampling weighing session has been completed, the analyst
 should prepare filters for sampling. The following steps should be followed:

 1.  Check the cassettes and the backing screens for fractures, cracks, chips, evidence of wear, or
     contamination. Clean the cassettes  prior to use and replace as necessary. The best practice
     is to immerse the cassettes, disassembled, into a sonicator bath with deionized water and
     detergent for 30 minutes and allow to air dry. Care must be taken in selecting a suitable
     detergent that will not leave residue on the cassettes  or screens. Residue on the cassettes
     will  contaminate the filters and produce elevated field and trip blanks.

      NOTE: Never use heat to dry cassettes. Heat causes the cassettes to warp, which
      will create field sampler problems.

 2.   Install each filter in a cassette, and  put the filter/cassette assembly into a protective
     container for transport to the monitoring site (see Figure 4.2). It is important to load the
     filters into  cassettes and ship quickly  because the filters will expire 30 days after pre-
     weighing, Attach a label (or COC)  with the filter identification number to the outside of the
     protective container. Prepare several  extra filters in case a filter is invalidated during the
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                       January 2016    Page 18 of 23

     installation process.

 3.  If the filters will be mailed, the field operator should be supplied with a cooler or other
     reinforced box that, in addition to the protective sample container, can protect exposed
     filters during their shipment back to the analytical laboratory, as well as keep filters at the
     temperature acceptance conditions discussed in Section 10.7 below.

 10.7   Post-Sampling Documentation and Filter Inspection

 All filter shipments returning to the laboratory from the field must be assessed to determine the
 time required for filter post-weighing. Upon receipt of the sample from the field, the analyst
 should follow these steps:

 1.     Examine the COC and/or field data sheet. Determine whether all data needed to ensure
       sample custody, verify sample validity, and to calculate mass concentration (e.g., average
       flow rate, ambient temperature and barometric pressure, and elapsed time) are provided.
       If data are missing or unobtainable from a field operator, or if a sampler malfunction is
       evident, record this information on the laboratory data form.

 2.     Inspect the samples for damage (see Section 10.3; damage may also include scratches in
       the filter, fingerprints, bugs, etc).  The analyst should document on the COC, the
       laboratory notebook, and/or the laboratory database if any samples arrived damaged. The
       analyst should document and thoroughly describe the type and extent of the damage. This
       information will be needed  by the monitoring organization's QA Coordinator (or decision
       maker) when making the final determination of data validity. Non-uniform deposits on
       the filters do not necessarily invalidate a sample, but should be documented. It is
       recommended that the post-weighing procedure be completed on all samples (even those
       that are damaged). The analyst should appropriately flag any  samples that have data
       integrity issues rather than not weighing them at all.

3.     Determine when the filters must be weighed.  This is determined by comparing the
       shipping receipt temperature to the average ambient temperature from the sampling run
       and applying the post-weighing criteria in 40 CFRPart 50, Appendix L, Section 8.3.6.
       Also, samples must always  be protected from temperatures exceeding 25°C during
       retrieval and shipping, pursuant to 40 CFR, Part 50, Appendix L, Section 10.13. Simply
       put, if the filters, from collection from the sampler through shipping to the weighing
       laboratory, are maintained at a temperature below the average ambient temperature  from
       the run and protected from  temperatures above 25°C, then the analyst must weigh the
       filters within 30 days. If the samples are received above the average sampling
       temperature from the run and below 25°C, then the analyst must weigh the filters within
       10 days.

       The best practice is to use ice substitutes with a min/max thermometer when collecting
       filters from the sampler, and then shipping the filters to the laboratory, to ensure the
       filters remain below the average ambient temperature of the sampling run. A
       thermometer is necessary to give  a quantitative record of the temperature to determine
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis
January 2016    Page 19 of 23
       which weighing time frame to follow. If filters are shipped cold with ice substitutes, a
       thermometer without min/max capability may be acceptable. However, EPA discourages
       shipping filters without a source of cooling. In the event that ice substitutes are not
       utilized, the cooler would need to contain a temperature logging device in order to track
       the temperature from sample collection to laboratory receipt.

       40 CFRPart 50, Appendix L, Section 8.3.6 includes a provision for cold weather
       sampling that dictates where the average  sampling temperature is below 4°C, the samples
       must be maintained from sample retrieval to the laboratory at less than 4°C to allow for
       the 30-day post-weigh time. If these filters collected in cold weather exceed 4°C during
       collection, shipping, or storage, but stay below 25°C, the analyst has only 10 days to
       weigh the filters.

       The following flowchart (Figure 10.6) may be used to assist the laboratory analyst in
       determining post-weighing time frames. In most cases, all filters arriving at the laboratory
       should be post-weighed. While the laboratory analyst may have determined using the
       flow chart or visual inspection that a filter will be invalid, the monitoring organization's
       QA Coordinator (or decision maker) should make the final determination of validity.
       However, it is the laboratory analyst's responsibility to clearly flag the filter  so that the
       QA Coordinator can make an accurate assessment of validity.
                       Yes
                                                   Start
                                                    Ave
                                                   Sample
                                                   Temp >
                                                  Monitoring
                                                 organization has
                                               violated filter temp.
                                                  regulations
                                                   Monitoring
                                                organization has 10
                                                days to weigh from
                                                  end of sample
                                                    period
                                                   Monitoring
                                                organization has 30
                                                days to weigh from
                                                  end of sample
                                                    period
        Samples
        arrive at <
          Ave
        Sample
         Temp
                                                                          No
                                                                                  Yes
      Figure 10.6 Filter sample holding time decision flowchart
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                       January 2016     Page 20 of 23

 3.     All sample shipments that were shipped cold and were, in fact, received cold should be
       immediately refrigerated at 1-4°C until several hours before the start of the filter
       conditioning period. When the analyst is ready to begin the filter conditioning process,
       the shipment should be removed from the refrigerator and allowed to warm to the filter
       conditioning environment's temperature before opening, in order to preclude water
       condensation on the sample filter(s) contained within. Remove the samples from the
       protective container and examine the container. If particulate matter or debris is found in
       the protective container after the filter has been removed, record that the sample has been
       flagged as questionable and state the reason as a remark at the bottom of the  laboratory
       data form. Save the filter for inspection. Notify the appropriate personnel.

 4.     Match the filter identification number with the correct laboratory data form on which the
       original microbalance number, filter number, pre-sampling filter weight, and other
       information were inscribed. Group filters according to the microbalance used to
       determine their initial tare weights. Initial separation of filters in this way eliminates the
       risk of a measurement error that could result from the use of different microbalances for
       pre- and post-sampling weighings.

 5.   Remove the filter from both the protective container and the filter cassette. Some cassettes
     may require special tools to disassemble them.

     NOTE: The stainless steel screen is sharp. Be very careful when removing the filter from
     the cassette. If the filter sticks to the stainless steel screen,  carefully lift the filter without
     tearing the Teflon®.  Sticking is generally an indicator of a poor screen cleaning process.

     Remove the cassette and stainless  steel backing screen to another area for cleaning. Be
     careful not to touch or otherwise disturb the filter and its contents. Transfer the filter to a
     filter-handling container labeled with the corresponding filter number. Place the used filter
     in the container "dirty-side" up. The filter should be handled with clean, smooth forceps
     and should not be touched by hands. If any damage occurs to the filter resulting from
     removing it from the cassette, the damage should be documented.

 6.   Transfer the filter in its filter-handling container to the filter conditioning environment.

 7.   Allow the filter to condition for at least 24 hours.

 10.8   Post-Sampling Filter Weighing (Gross Weight)

 Both the pre- and post-sampling filter weighings should be performed on the same  analytical
 balance. Different analysts can perform the pre- and post-sampling filter weighings as long as
 the appropriate standard operating procedures (SOPs) have been followed and the working
 standard and duplicate measurements are within specifications. Use an effective technique to
 neutralize static charges on the filter. The post-sampling conditioning and weighing shall be
 completed within 240 hours (10 days) after the end of the sampling period, unless the filter is
 maintained in the manner described in Section 10.7 that would allow for weighing within 30
 days from sample retrieval.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0  Filter Preparation and Analysis                       January 2016    Page 21 of 23

 Follow these steps during post-sampling filter weighing:

 1.   Group filters in numerical order according to the microbalance used for pre-weighing and
     by their filter numbers. Reweigh each filter on the same microbalance on which its pre-
     sampling weight was obtained.

 2.   Repeat Steps 1 through 14 in Section 10.6, noting the following differences.

     a.   At least one laboratory blank from the same pre-weigh batch, and any available field
         blank, should be weighed. If the pre- and post-sampling weights for the laboratory
         blanks disagree by more than 15 ug, repeat the measurements. If the pre- and post-
         sampling weights for the field blanks disagree by more than 30 ug, repeat the
         measurements. If the two measurements still disagree, troubleshoot and take
         appropriate corrective action, as specified in the monitoring organization's QAPP.
         Measurements for sampled filters should not be corrected to account for blank
         measurements.

         NOTE:  Occasional high blank values should not cause the automatic invalidation of
         sampled filters that were measured during the  same weighing session. Instead, such
         values should trigger troubleshooting and corrective action to reduce blank values to
         acceptable levels.  However, repeated,  excessively high blank values could indicate a
         significant underlying issue in the laboratory, such as improper grounding, which
         could jeopardize the quality of the data, if not resolved. Control-charting laboratory
         blank values is recommended.

     b.   One routine filter should be reweighed as a batch duplicate. For this post-sampling
         weigh session, however, the filter chosen should be an exposed (sampled) filter. The
         acceptance criterion remains ±15 ug. If the duplicate measurement disagrees from the
         original measurement by more than 15 ug, reweigh the filter. If the measurements still
         disagree, troubleshoot and take appropriate corrective action, which may include (1)
         reweighing all of the previously weighed filters, (2) reweighing the working standards,
         or (3) having a service technician repair the microbalance. The analyst should not
         attempt to repair the microbalance.

 3.   If the filter(s) will receive further analysis, return it to the filter-handling container and note
     on the container and the laboratory data form that additional analyses are required. Transfer
     the filter to the laboratory responsible for performing the additional analyses.

 10.9   Calculation of Net Mass Filter Loading

 A filter's post-sampling mass minus its pre-sampling mass is the net mass loading for that filter.
 Record this value on the laboratory  data form. Refer to Section 12.0, "Calculations, Validations,
 and Reporting of PM2.5 Monitoring  Data,"  for the calculations required to compute and report
 ambient PM2.5 concentrations in ug/m3. Measurements for sampled filters should not be
 corrected to account for blank measurements.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0 Filter Preparation and Analysis                         January 2016    Page 22 of 23

10.10  Internal Quality Control

See Table 9-1 for a summary of the internal QC requirements for the pre- and post-sampling
weighing sessions.

The laboratory data management system (DMS) should contain weighing session QC data and
may contain supplemental QC documentation such as: the microbalance calibration and
maintenance information, routine internal QC checks of mass reference standards and laboratory
and field filter blanks, and external quality assurance (QA) audits. The DMS should be designed
to plot QC data (results of blanks, duplicate weigh sessions, and the weigh room's environmental
conditions) on control charts so that long-term trends can be identified. It is recommended that
QC charts be maintained on each microbalance and routinely plotted and reviewed. These charts
may indicate instrument malfunctions, procedural problems, and/or poor quality supplies or
materials,  before significant data impacts are realized.

If more than one microbalance is used in the laboratory, the pre- and post-sampling
measurements of the filter should be made on the same microbalance.

It is also recommended that someone other than the routine analyst reweigh a portion of the
filters from a manual  weigh session (batch) as an additional QC check. This duplicate weighing
by an independent person provides further confidence in the results of the weighing session. If
this best practice recommendation is implemented, it is suggested that 7% of the filters in the
batch be reweighed, the results of which documented in the laboratory QC logbooks and/or
database.

It is further recommended that the temperature and relative humidity devices in the weighing
room be verified on a routine basis. Verification of these devices can be performed by the
analyst,  although an independent (QA) staff member is suggested. The verification should be
performed using independent, NIST-traceable temperature and relative humidity standards (see
Section  4.3.9). The results of the verification checks should be recorded on a data form, in the
laboratory logbook, and/or laboratory QC database. Temperature results should fall within ±2°C
of the audit standard; relative humidity values should fall within ±2% of the audit standard. If the
verification results  do not fall within these guidelines, the temperature and/or relative humidity
devices  may need to be returned to the vendor for recalibration/recertification. See Section 11.2.8
for additional information.

 10.11   Filter Archive Requirements

 Air pollution control agencies must archive the exposed, final-weighed PIVh.s filters from all
 SLAMS sites. 40 CFRPart 58.16(f) specifically states:

      The state,  or where applicable, local agency shall archive all PM2.5, PM10, andPMlO-
      2.5 filters from manual low-volume samplers (samplers having flow rates less than 200
      liters/minute) from all SLAMS sites for a minimum period of 5 years after collection.
      These filters shall be made available for supplemental analyses, including destructive
      analyses if necessary, at the request of EPA or to provide information to state and local
      agencies on particulate matter composition. Other Federal agencies may request access
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.10.0 Filter Preparation and Analysis                         January 2016     Page 23 of 23

      to filters for purposes of supporting air quality management or community health—such
      as biological assay—through the applicable EPA Regional Administrator. The filters shall
      be archived according to procedures approved by the Administrator, which shall include
      cold storage of filters after post-sampling laboratory analyses for at least 12 months
      following field sampling. The EPA recommends that paniculate matter filters be archived
      for longer periods, especially for key sites in making NAAQS-related decisions or for
      supporting health-related air pollution studies.

Cold storage involves refrigerating exposed sample filters; archived filters should not be frozen.
After removal from cold storage, the filters should be relocated to a clean, dry area, where they
will be protected from light, vibrations, and dust sources.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
                      January 2016     Page 1 of 15
 11.0  Performance Evaluation Procedures

 11.1   Overview

 This section summarizes the performance evaluation (PE) procedures for assessing
 measurement uncertainty, precision, and bias for reference or Class I equivalent methods for
 PM2.5 sampling at State and local air monitoring stations (SLAMS) sites. These requirements
 are specified in 40 CFR Part 58, Appendix A. This section also gives suggestions for
 components of performance and systems audits of samplers, balances,  and operational
 procedures that Primary Quality Assurance Organizations (PQAO) may wish to conduct, in
 addition to those to be provided by the U.S. Environmental Protection  Agency (EPA) Regional
 Office.

 Table 11-1 consolidates the required and recommended PE components and lists the personnel
 responsible for conducting  and reporting the results, the frequency at which the audits or checks
 are to be performed, and the performance goal.
Table 11-1.  Systems and Performance Evaluation Components for Manual Method PIVh.s
          Samplers24


1.

Component
Sampler flow rate audit, (a) Sampler
display versus audit device reading, (b)
Audit device reading versus designated
flow rate (required).
Conducted
by
PQAO

Frequency
Every 182 days
and twice a
year
Performance
goals
(a) ±4 percent
(b) ±5 percent
 2.  Precision via collocated sampler at 15
     percent of sampling sites for each method
     designation of a primary sampler
     (required).

 3.  Federal Reference Method (FRM) PE via 1
     to 2 days collocation of an FRM sampler at
     5 or 8 sampling sites each year (required).

 4.  Balance systems and performance audits
     (recommended).
  5.  Sampler performance audits (temperature,
     pressure, flow rate) (recommended, based
     on quality assurance project plan [QAPP]).

  6.  Systems audit (required).
  7.  Systems audit (recommended).
PQAO
PQAO or EPA
Regional Offices


PQAO
PQAO
EPA Regional
Offices

PQAO or State
Every 6 days
5- 8 PEP
values a year


Every 365 days
and once a
year

Per QAPP
Once every 3
years

Soon after
network startup
& periodically
thereafter
10 percent
coefficient of
variation (CV) for
total precision

±10 percent for total
bias
Meet QAPP goals;
agreement within
Per QAPP
Meet CFR and
QAPP goals

Meet CFR and
QAPP goals
24 See Validation Template on AMTIC for any revisions to acceptance criteria in this table
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January2016    Page2of15

 For the PE components listed in Table 11-1 that are conducted by the SLT agency or PQAO, it is
 expected that the agency or organization will include details in its QAPP and standard operating
 procedures (SOPs) that describe:

   •   Conducting and reporting of the required semi-annual (i.e., 182 days) sampler flow rate
       audits
   •   Conducting and reporting of data to AQS from the required collocated samplers to
       estimate precision
   •   Any additional systems or performance evaluations it wishes to conduct.

 In general, any PE or audit, whether conducted by personnel outside or within the monitoring
 organization, should be conducted under the following guidelines:

   •   Without special preparation or adjustment of the system to be evaluated.

   •   By an individual with a thorough knowledge of the instrument or process being evaluated,
       but not by the routine operator.

   •   With accurate, calibrated/certified, National Institute of Standards and Technology
       (NIST)-traceable transfer standards that are completely independent of those used for
       routine calibration and QC  flow checks (although both calibration standards and audit
       standards may be referenced to the same primary standard for flow rate, volume,
       temperature, pressure, or mass).

   •   With complete documentation of audit information. This documentation includes, but is
       not limited to, the types of instruments evaluated, the audit transfer standards and transfer
       standard traceability, the instrument model and serial numbers, the calibration
       information, and the collected audit  data.

 Overviews of the procedures for conducting performance and systems evaluations of the PIVfo.s
 collection and measurement processes are provided in these sections:

   •   Section 11.2, "Performance Evaluations" (sample flow rate audit, precision assessment,
       FRM PEP, balance accuracy, and so forth); and,

   •   Section 11.3, "Systems Audits" (EPA Regional Office audits and internal audits conducted
       by the monitoring  organization or PQAO).

 11.2   Performance Evaluations

 This section presents performance audit procedures specific to reference or Class I equivalent
 PM2.5 samplers that operate on the inertial  impaction principle at a sample flow rate of 16.67
 L/min. Procedures for auditing samplers that operate according to other principles (e.g., open-
 path optical monitors) are not discussed here. 40 CFR Part 58, Appendix A requires that PQAOs
 assess, on an -182 day basis (and twice a year), a flow rate  audit of each primary (data-reporting)
 PM2.5 sampler used in their SLAMS monitoring network and report this data to AQS.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January 2016     Page 3 of 15
 Appendix A of 40 CFR Part 58 also requires PQAOs to collocate 15 percent of its samplers,
 based on method designation (method codes25). Furthermore, the PQAOs are required to assess
 total measurement system bias by making side-by-side comparisons between the site's primary
 sampler PM2.5 concentration results and those of an FRM Performance Evaluation Program
 (PEP) sampler that is temporarily installed adjacent to the site sampler. This operation may be
 implemented by the EPA Regional Office.

 This section also offers guidance to SLTs or PQAOs that may wish to incorporate internal
 audits into their PIVh.s monitoring program. Internal performance audits are recommended for
 sampler flow rate, temperature, and barometric pressure. Balance audits are also suggested.
 Monitoring organizations may elect to include an internal audit of the operational aspects of
 PM2.5 samplers as part of their overall QAPP. The person conducting these audits should be
 someone other than the regular site operator. It is also important that the audit devices for flow
 rate, temperature, and barometric pressure not be the ones used for regular site checks and
 calibrations. However, the audit transfer standards may be traceable to the same primary
 standards as the calibration standards.

 11.2.1    Sampler Flow Rate Audit

 The calibration relationship between the flow rate (in actual L/min) indicated by the flow rate
 standard and the sampler's flow rate measurement device should be established or verified in
 accordance with the sampler's operations or instruction manual prior to the audit; however, no
 adjustments may be made to the sampler's flow rate sensors or controls before the audit is
 performed.  A suggested audit data recording form is shown in Figure 11.1. The example audit
 data sheet shown in Figure 11.1 should be styled to include additional information as necessary.
 In conducting the sampler flow rate audit (or any audit for that matter), the auditor should
 consult the manufacturer's operations manual for sampler-specific procedures, suggestions, and
 precautions.

 The audit of the sampler's flow rate consists of one measurement made at the sampler's
 operational flow rate. The flow rate transfer standard used for the audit must not be the same
 flow rate standard used to verify  or calibrate the sampler, but may have been made traceable to
 the same primary standard for flow rate or volume. The sampler's operations or instruction
 manual should provide specific instructions for using the instrument's flow rate readout or
 display. The audit is performed by the operator or by personnel from the quality assurance (QA)
 unit of the monitoring organization as follows:

 1.   Install  a flow rate audit filter/cassette in the filter  holder of the sampler. Filters used for
     flow rate audits should not be used further for sampling. If the network's sampling
     schedule is on a daily basis,  or if the loss of a sampling day cannot be accommodated, a
     flow rate audit may be conducted during a sampling event provided that:

   •   The active sampling filter is removed and protected from  contamination and damage by
       storing it in a clean container between uses;
25 Method codes can be found in AQS


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                      January 2016    Page 4 of 15
   •   A flow rate audit filter is used for the audit; and,

   •   The active sampling period is not interrupted for more than 1 hour in order to achieve a
       minimum overall sampling period of 23 hours.

 2.   Complete the required information at the top of the sampler audit data sheet (Figure 11.1).

 3.   Generate flow on the sampler and observe the "as found" value (in L/min). Record this
     value on the audit data sheet and/or in the sampler logbook.

 4.  Perform an external leak check as described in Section 7.4.3. This value also verifies the
    "as found" condition of the sampler. Record the value on the audit data sheet and/or in
    the sampler logbook.

 5.  Perform the flow rate audit. Connect the flow rate transfer standard to the sampler
    downtube. If using the flow audit adapter with flexible tubing to connect the flow rate
    standard, be careful not to crimp the tubing. If the transfer standard is  electronic (such as an
    electronic manometer coupled to an orifice device), allow the audit standard sufficient
    warm-up time, based on one's prior experience with the device, for readings to stabilize.

 6.  Observe the flow rate indicated by the PIVb.s sampler's data display or other readout device.
    Record this value and the corresponding flow rate indicated by the audit device on the audit
    sheet and/or in the sampler logbook. If units other than actual flow are displayed by the
    audit device, the ambient temperature and barometric pressure will have to be independently
    determined and calculations will have to be made to express results at ambient conditions.
    See Section 11.2.1.1 for calculation methods. Also, compare the audit flow rate with the
    designated flow rate, 16.67 L/min.

 7.  Determine the percent differences in flow rate. Record all data for eventual  submission to
    the EPA AQS database.

 8.  Following the measurement with the flow rate standard, remove the adapter/standard and
    reinstall the sampler inlet. From the sampler's display, read the sampler's operating flow
    rate again (in L/min) with the flow rate filter in place. Ensure the flow rate has not
    significantly changed from that observed in Step 3 above, which could indicate an issue
    developed within the sampling train resulting from the audit process. Remove the flow rate
    audit filter. Do not use it for further sampling; it may be used several more times for flow
    rate audits of the same model sampler,  provided it is carefully stored between uses to
    prevent contamination or damage. Stop using this filter when a buildup of particles on the
    filter surface is visible.

 9.  Replace the sample  filter if it was removed in the midst of a sampling  event. Otherwise,
    return the sampler to its pre-audit condition.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January 2016     Page 5 of 15

 11.2.1.1 Flow Rate Audit Data Calculations

 PM2.5 flow rate audit results should be reported at ambient conditions and should not be
 corrected to standard conditions (e.g., 25 °C, 760 mmHg). However, for some types of audit
 devices, it may be necessary to correct the indicated audit flow rates to ambient conditions.
 Equation (11-1) may be used to make this  correction. The following correction should be
 applied only if the audit device reports flow rate already corrected to standard conditions (e.g.,
 mass flowmeters). Be sure to verify the standard conditions of temperature and pressure (Tstd
 and Pstd, respectively) applicable to the audit device.

                                 Qa = QstA(Ta/Pa) (Pstd/Tstd)                            (11-1)

 where

            Qa = flow rate at actual conditions, L/min
           Qstd = flow rate reported at standard temperature and pressure  (e.g., 25 °C, 298 K;
                  760 mmHg or 101 kPa), L/min
             Ta = ambient temperature, K
            Pa = ambient barometric pressure, mmHg or kPa
       Pstd, Tstd = standard barometric pressure and temperature, respectively.

 Determine the percentage difference between the sampler-indicated flow rates and the audit-
 measured flow rates using Equation (11-2). Record the result on the audit data sheet. Any
 deviation greater than ±2 percent may require recalibration.

                   % difference =  100 x (ga(sampler) - £>a(audit))/ga(audit)             (JJ-2)

 11.2.2  Temperature Audit Procedures

 Ambient temperature, filter temperature, and ambient pressure sensors on the PM2.5 sampler
 may be audited in conjunction with the flow rate audit. Make no adjustments to the sampler
 prior to recording all audit results.  Record the audit data on a form similar  to the one shown in
 Figure 11.1.

 A one-point temperature audit is sufficient. The temperature standard used for the audit must
 not be the same standard used to verify or  calibrate the sampler, but may have been made
 traceable to the same primary standard for temperature.  If the audit fails (assuming the audit
 device is accurate), it is documented and reported and calibration of the samplers' temperature
 sensors may be warranted. Details  of temperature calibration procedures are found in Section
 6.4, "Calibration Procedures," of this document and can be adapted for audit purposes. Record
 the stabilized temperature reading  and report the results on a data sheet such as the one shown
 in Figure 11.1. Any deviation greater than 2 °C should be reported for corrective action.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
              January 2016    Page 6 of 15
                                 PM2.5 AUDIT FORM
                            (Flow Rate, Temperature, and Pressure)
              INSTRUMENT
                  SITE
 Manufacturer:
 Model No.:
 Siv./Ser. No.:
             FLOW DEVICE
 Device:
 Serial Number:
 Cert Date:
AQSSiteNo.:
Location:
Operator:

Auditor:  	
Audit Date:
         TEMPERATURE PROBE
 Cert Date:
              BAROMETER
 Cert Date:
FLOW RATE

Flowate
Set Flowrate
Siitial Leak Test
Final Leak Test
Sampler Flow
Reading (LPM)

16.67
mmHg
mmHg
Audit Flow
(LPM)

Comparison to Audit

Operator Leak Test
% Difference
4% to Audit Device; 5% to Target



mmHg
TEMPERATURE
Temperature
Range

Zero (ice bath)
Ambient
High
Ambient
Sensor
Sampler



Ambient
Sensor
Audit



Degree
difference
2° Limit



Filter
Sensor
Sampler



Filter
Sensor
Audit



Degree
difference
2° limit



Comp
Sensor
Sampler



Comp
Sensor
Audit



Degree
difference
2° limit



PRESSURE

Barometric
Pres s ure (mm Hg)
Sampler

Barometric
Pressure (mmHg)
Audit

mmHg
(10 mmHg Limit)
Difference





Comments :
TIME
Sampler

Audit

Difference






              Verified By:
                     Date:
 Figure 11.1 Example PM2.5 Sampler Audit Data Sheet
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January 2016    Page 7 of 15

 11.2.3  Barometric Pressure Audit Procedures

 Ambient pressure is audited using a suitably calibrated sensor at ambient conditions. Refer to
 Section 6.5 of this document for information on establishing traceability and the use of portable
 barometers, and Section 7 for the audit procedure, which would be very similar to the
 verification procedure but with the use of an independent device.  Record all findings on an
 audit data form. Any deviation greater than 10 mmHg should be reported for corrective action.

 11.2.4 Time Audit

 For a time audit of the sampler, one can use a cellular phone that can access the official NIST
 Clock at www.fime.gov (i.e., atomic clock), or has been checked against this clock and is within
 30 seconds of it.  Most celluar phones are synced to the NIST Time, so it will be rare that the
 celluar phone will be off. The sampler's time must be within 60 seconds of the NIST time.

 11.2.5  Assessment of Precision Using Collocated Samplers

 Collocated sampler results, where the collocated sampler is not necessarily a reference method
 sampler but a sampler of the same designated method as  the primary sampler,  are used to assess
 measurement system precision according to the schedule and procedure specified in 40 CFR
 Part 58, Appendix A, Section 3.2.3. The regulations require that 15 percent of a PQAO's
 primary samplers have another sampler (termed the "QC monitor" in CFR) collocated at the
 same site and collect a sample on at least an every sixth day schedule, such that the primary and
 collocated samplers capture samples from all days of the week throughout the year.  The site
 locations for the collocated samplers are decided by choosing those sites that are most likely  to
 have higher concentrations of PIVfo.s or that have been found to be in violation of the National
 Ambient Air Quality Standards (NAAQS) during initial studies. Refer to 40 CFR Part 58,
 Appendix A, Section 3.2.3 for details. The collocated sampler measurements will be used by
 EPA to calculate quarterly and annual precision estimates for each primary sampler and for each
 designated method employed by each PQAO. A data quality objective (DQO) of 10 percent CV
 or better has been established for the operational precision of PIVb.s monitoring data.

 11.2.5.1 Procedure for Collocated Measurements

 The collocated QA sampler must be kept clean, serviced, and inspected frequently for proper
 operation,  and the calibration of its flow rate, temperature, and pressure measurement systems
 must be checked and verified at the same intervals as the primary sampler.

 The station's primary sampler and the collocated (audit)  sampler must be located at least 1
 meter but no more than 4 meters apart, and their inlets should be at the same height  above
 ground and within 1 meter of each other as measured in the vertical direction.  Space either
 sampler at least 2 meters away from a high-volume sampler. The samplers must be calibrated
 and operated according to their operations manuals, and the samples must be stored, shipped,
 and analyzed identically, as described in the sampler's operation or instruction manual and in
 general accordance with the guidance in this document and in the monitoring organization's
 QAPP and SOPs.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January 2016     Page 8 of 15

 The concentrations determined from the site's primary sampler and from the collocated sampler
 are to be reported to EPA as directed in 40 CFR Part 58, Appendix A, Section 4.

    NOTE: All collocated measurements must be reported, even those that are considered
    invalid because of identified malfunctions or other problems that occurred during or
    following the sample collection period. (If invalid, the data should be reported with an
    AQS null value code.) Although all measurements must be reported, only those sample
    pairs having PM2.5 concentrations above 3 ug/m3 will be selected for use in precision
    calculations, because at low concentrations agreement between the measurements of
    collocated samplers may be relatively poor.

 11.2.6   Federal Reference Method  Performance Evaluation  Program

 Because the size and volatility of the particles making up ambient particulate matter vary over a
 wide range and the mass concentration of particles varies with particle size, it is difficult to
 define the accuracy of PM2.5 measurements in an absolute sense. The accuracy of PM2.5 mea-
 surements is therefore defined in a relative sense, referenced to measurements provided by the
 FRM. Accuracy is defined as the degree of agreement between a subject field PM2.5 sampler
 (the primary sampler) and a collocated FRM Performance Evaluation Program (PEP) sampler
 operating simultaneously at the monitoring site and totally independent of the site operations
 with respect to operator, sampler, sample handling, and weight determination. Audit results
 include both random (precision) and systematic (bias) errors. The PM2.5 PEP is described in 40
 CFR Part 58, Appendix A, Section 3.2.4.

 The EPA Regional Offices are responsible for conducting the FRM PEP to assess total
 measurement system bias. The PQAO  is responsible for assisting FRM PEP personnel by
 coordinating schedules and assisting in the setup of the FRM sampler, which will be on site for
 approximately 2 days.

Primary quality assurance organizations with 5 or less PM25 monitoring sites are required to have
5 valid audits per year, distributed across the  4 quarters; PQAOs with greater than 5  sites are
required to have 8 valid audits per year, distributed across the 4 quarters. The EPA requires:

     •   One hundred (100) percent completeness (meaning whatever it takes to get  5 or 8 valid
        samples); and
     •   All  samplers subject to an audit within 6 years.

 Details of the schedule for performing collocated measurements with the FRM sampler and
 guidelines for determining the initial deployment of FRM PEP samplers based on sites likely  to
 be in violation of the NAAQS are provided in 40 CFR Part 58, Appendix A, Section 3.2.4. The
 results  from the primary sampler and the FRM PEP sampler are used by EPA to calculate the
 bias of the primary sampler on an annual basis, and the bias of a PQAO on an annual and three-
 year basis.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                     January 2016     Page 9 of 15

 11.2.7   Balance Accuracy Assessment

 Each monitoring organization should conduct an internal accuracy assessment of each
 microbalance on an annual basis. A performance audit of the microbalances used to weigh
 PM2.5 filters requires the use of an independent set of American Society for Testing and
 Materials (ASTM) Class I standard weights (tolerance < 0.010 mg), traceable to NIST. These
 weights must not be the same ones used as working standards for the day-to-day operation of
 the microbalance,  but may be traceable to the same primary standard used to trace the working
 standards. Because microbalances are extremely delicate instruments and should not be
 operated by inexperienced personnel, it is recommended that the PE of the filter-weighing
 process be done in cooperation with the laboratory personnel. The person normally performing
 the weighings for PM2.5 monitoring should assist the auditor by preparing the balance as if a
 series of filter weighings were to be done. Record all data on an audit data sheet and/or in the
 laboratory's balance notebook. The balance display  should agree with the designated value of
 the audit weight (i.e., its certified weight/mass) to within twice its uncertainly (or at least ±0.003
 mg).

 Many laboratories maintain an agreement with a service representative to conduct regular
 servicing of the balances. It may be instructive to conduct a performance audit prior to the
 periodic servicing and again immediately after the servicing.

 11.2.8   Temperature  and  Relative Humidity Logger Audit Procedure

 The temperature and RH data logger of the environmental control system for the PM2.5
 weighing room is an integral part of ensuring 40 CFR Part 50 Appendix L weighing  criteria are
 met and should be audited on a routine frequency. Dataloggers may fail in different ways. A
 comparison of all numbers, logged and real-time, is essential to ensure that the loggers are both
 functioning properly. An independent audit standard or logger should be used to compare
 against the laboratory temperature and RH unit every 90 days; however, more frequently is
 recommended. The standard used for the audit must be at least as accurate as the weighing
 room's logger. The acceptance criteria for the temperature and RH audit is ±2° C for
 temperature and ±2% for RH.

 The audit standard should be placed in the weighing room and allowed to equilibrate for at least
 30 minutes prior to conducting the audit. After the audit standard has equilibrated to room
 conditions, record the audit standard temperature and RH readings and the temperature and RH
 from the laboratory logging system every five minutes for 30 minutes. Calculate the  differences
 for both temperature and RH for all points. Calculate and compare the means of the audit device
 and the logger. If the comparison is outside of the acceptance criteria, troubleshoot and repeat
 the process. If the  problem persists, cease weighing, replace the laboratory logger, and
 investigate the impact on the filter weighing data.

 Figure  11.2 is an example form for performing a temperature and relative humidity logger audit.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
January 2016     Page 10 of 15
                        Temperature and Humidity Verification Form
Dale
Laboratory Analyst
Laboratory Analyst's Stgnalure

Audit Standard
Model and Senal Nunber
Calibration Due Date
WewjWng Room Logger
Mode! araj S€nal Numder
Calibration Dye Date






Observation
1
2
3
4
5
6
Means
Time





.
Audit Standard
Temp "C






SDIViOi
%RH






SDIVVQI
Weigning Roan Logger
Logged Data
Temp °C






SQIV/QI
%RH






sQIV/Qt
Real-time Date
Temp°C






*DIV/OI
%RH






oorwoi
Nofrer&al Differences
Logged Data
Degrees *C
00
00
00
00
00
00
00
Percent RH
00
00
00
00
00
00
00
Real-tune Data
Degrees *C
00
00
00
00
00
00
00
Percent RH
00
00
00
00
00
00
00
Enter Acceptance Criteria Temp +/• * RH +(• % ore* one
Pass Fail

If verification fails, note troubleshooting action and resolution:
 Figure 11.2 Temperature and humidity verification form

 11.3   Systems Audits

 Periodic systems audits of PM2.5 programs performed by the monitoring organization's QA staff
 are strongly recommended. Such audits are particularly important at the startup of a new
 monitoring effort and can help recognize and pinpoint problems before significant amounts of
 data are affected resulting in data invalidation. Loss of data can affect the ability to make
 NAAQS decisions.

 Pursuant to 40 CFR 58, Appendix A, Section 2.5, the EPA Regional Offices are charged with
 conducting a systems audit once every 3 years. These audits are designed to assess the entire
 measurement system and data management activities with regard to PM2.5 monitoring networks.
 These activities may include the following:

   •   Initial equilibration, weighing, and transportation of the filters to the sampler
   •   Site selection criteria assessment
   •   Equipment installation
   •   Site security
   •   Equipment maintenance
   •   Calibration procedures
   •   Handling and placement of the filters
   •   Operation of the sampler and sample collection
   •   Removal, handling, and transportation of the filter from the sampler to the laboratory
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures                    January 2016     Page 11 of 15

   •   Weighing, storage, and archival of the sampled filter
   •   Data analysis and reporting.

 EPA specifies that QA and QC programs follow the requirements for QAPPs contained in the
 EPA document Requirements for Quality Assurance Project Plans for Environmental Data
 Operations, EPA QA/R-5 (EPA 1994). Auditors should develop a checklist based on these
 guidelines and on the specific information regarding PM2.5 monitoring in this guidance
 document.

 Effective system auditing requires the auditor to observe the auditee's entire sample acquisition
 and analysis system and the associated data processing operations. The locations audited
 include the sampling site, laboratory, the shipping and receiving area where field samples are
 received and logged in, the sample archival area, and the data processing areas including the
 laboratory DMS.

 Data about specific samples obtained during on-site performance audits may be useful during
 the systems audit. Tracking specific samples and the associated data completely through the
 system (sample acquisition, shipping,  receiving, logging in, weighing, archiving, data
 processing, and reporting) ensures that every significant operation is examined. This approach
 also allows the auditor to look at the "interfaces" between different operations—the interactions
 between field and laboratory tasks—and not just the individual tasks in isolation. Detailed and
 summary reports and data submitted to the EPA database (AQS) for the audited period should
 also be reviewed during the audit.

 Effective preparation for a systems audit requires developing a checklist similar to the example
 shown in Figure 11.3. Auditors are encouraged to develop their own checklists based on their
 experience and as dictated by the individual audit situation.

 Periodic internal systems audits of PM2.5 programs conducted by the monitoring organization
 are strongly recommended, and should be included in the organization's QAPP. Such audits are
 particularly important at the startup of a new monitoring effort and can help recognize and
 pinpoint problems before  significant parts of data are affected. After start up, an annual systems
 audit of the PIVb.s program, including  the gravimetric laboratory, is encouraged. Loss of data
 could affect the monitoring organization's designation classification; conducting internal
 systems audits is a proactive approach to ensure the monitoring organization (PQAO) is
 continuously meeting regulatory requirements and producing quality data.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
                               January 2016     Page 12 of 15
   Site or Laboratory Location
   Auditor Name and Affiliation
Checklist for PM2.s Auditing
                    Site Designation
   Observer(s) Name and Affiliation
Audit Question
Y
N
N/A
Comment
On Site:
Are logbooks and required data sheets filled in promptly,
clearly, and completely?
Are filters handled with the necessary care and finesse to
avoid contamination and/or loss of material? Does the
operator keep the filter-handling area neat and clean?
Is the equipment sited properly? Are there any changes at
the site that might compromise original siting criteria (e.g.,
fast-growing trees or shrubs, new construction)?
Does the equipment appear to be well maintained and free
of dirt and debris, bird/animal/insect nests, excessive rust
and corrosion, etc.?
Are the walkways to the station and equipment kept free of
tall grass, weeds, and debris?
Is the station shelter (if any) clean and in good repair?
























Sampler Calibration and Maintenance:
Does the flow rate standard used for routine flow rate
calibration have its own certification traceable to a NIST
primary standard for volume of flow rate?




                                                                                     (continued)
Figure 11.3 Audit checklist
                                                 2.12-PM25 Reference/Class I Equivalent Methods

-------
 2.12.11.0 Performance Evaluation Procedures
January 2016    Page 13 of 15
Audit Question
Y
N
N/
A
Comment
Sampler Calibration and Maintenance (continued):
Is the calibration relationship for the flow rate standard (e.g.,
an equation, curve, or family of curves relating actual flow
rate [Qa] to the flow rate indicator reading) accurate to within
2 percent over the expected range of ambient temperatures
and pressures at which the flow rate standard may be used?
Verify by inspecting the original calibration data and
calculating the error in the calibration relationship for each
calibration point used to generate the relationship.
Is the flow rate standard used for routine sampler
calibration/verification recalibrated or reverified against a
NIST-traceable standard at least annually?








Balance Maintenance and Weighing Procedures:
Is the filter preparation and weighing area neat and clean?
Are logbooks kept up to date and properly filled in?
Is the balance on a service agreement for regular profes-
sional maintenance? If not, is there someone in the organ-
ization with the appropriate training to service the balance?
Does the analytical balance used to weight filters have a
readability of ±1 jig?
Is there a formal logbook or file for balance maintenance?
Are entries current?
Are regular (e.g., daily, when in use) calibration checks
made and recorded?
Are filters conditioned immediately before both the pre- and
post-sampling weighings? The following conditions are
specified in 40 CFR Part 58, Appendix L:




























Mean temperature: 20-23 °C.
Temperature control: ±2 °C over 24 hours.
Average humidity: 30 to 40 percent relative humidity (RH); however, where it can be shown that the mean
ambient RH during sampling is between 20 and 30 percent, conditioning is permissible at a mean RH within ±5
percent of the mean ambient RH during sampling.
Humidity control: ±5 percent RH over 24 hours.
Conditioning time: Not less than 24 hours.
Are new filters placed in the conditioning environment
immediately upon arrival and stored there until the pre-
sampling weighing?




Figure 11.3 Audit checklist (continued)
                                                  2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
January 2016    Page 14 of 15
Audit Question
Y
N
N/
A
Comment
Balance Maintenance and Weighing Procedures (continued):
Is the analytical balance located in the same controlled
environment in which the filters are conditioned?
Are the filters weighed immediately following the conditioning
period without intermediate or transient exposure to other
conditions or environments?
Are filters conditioned at the same environmental conditions
before both the pre- and post-sampling weighings?
Are both the pre- and post-sampling weighings performed on
the same analytical balance, using an effective technique to
neutralize static charges on the filter?
Are both weighings performed by the same analyst? (If not,
have results of the different analysts been compared
statistically?)
Are pre-sampling weighings done within 30 days of the
sampling periods in which filters are to be exposed? (Review
logbook to look for evidence of filters that exceed their time
span.)
If filters are stored at ambient temperature, are the post-
sampling conditions and weighings completed within 240
hours (10 days) after the end of the sample period?
If filters are stored at 4 °C or below during the entire time
between retrieval from the sampler and start of
conditioning, are the post-sampling conditioning and
weighings completed within 30 days after the end of the
sample period?
Are new field blank filters weighed along with the pre-
sampling (tare) weighing of each lot of PlVh.s filters?
Are QC field blank filters routinely used, observing the
following handling steps: transport to the sampling site,
installation in the sampler, retrieval from the sampler (without
sampling), and reweighing?
Are QC laboratory blank filters weighed along with the pre-
sampling (tare) weighing of each set of PlVh.s filters and
reweighed when the exposed filters are received from the
field? (These laboratory blank filters should remain in the
laboratory in protective containers during the field sampling
and should be reweighed as a QC check.)
Was the balance calibrated as specified by the manufacturer
at installation and recalibrated immediately prior to each
weighing session?
















































 Figure 11.3 Audit checklist (continued)
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.11.0 Performance Evaluation Procedures
January 2016    Page 15 of 15
Audit Question
Y
N
N/
A
Comment
Recordkeeping and Calculations:
Are logs and/or charts of the balance room temperature and
humidity on file?
Are records of shipments (incoming and outgoing)
maintained?
Are records of sample filter condition upon arrival at
laboratory kept (i.e., temperature)?
Are data management files in order?
Is there evidence that data validation, internal QA review,
and complete data reporting have occurred?
Is the personnel management structure sound?
























 Figure 11.3 Audit checklist (continued)
                                                2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
        PM2.5 Monitoring Data                                    January 2016    Page 1 of 7

 12.0  Calculations,  Validations, and Reporting of PIVfe.s Monitoring
        Data
 12.1   Overview

 This section presents the calculations required to compute and report ambient PIVfo.s concentra-
 tions. A summary of all calculation formulas is given in Table 12-1. The end result of these
 calculations is the integrated (average) PM2.5 mass concentration during the sampling period,
 expressed in units of ug/m3. To obtain this concentration, it is necessary to determine the mass
 of particles collected by the filter and the total volume of air that was sampled. The volumes
 used to calculate the PIVb.s concentration must always be expressed at the actual (local) ambient
 temperature and pressure used to acquire the sample (40 CFRPart 50.3).

 Accurate measurement and control of the sampler's flow rate are needed to determine the total
 sampled volume  and to ensure that the particles trapped by the filter are in the correct particle
 size range. Reference samplers and Class I equivalent samplers separate PIVh.s particles by
 inertia. Obtaining the correct particle size range by inertial separation requires maintaining the
 correct air velocity in the sampler's PM2.5 separator. This velocity is correct when the specified
 design flow rate is maintained.

 12.2   Calculations

 12.2.1    Sample Volume Calculations

 Both reference and equivalent method samplers are required to provide measurements of the
 total volume of air sampled (Fa), in m3 at the actual ambient temperatures and pressures during
 sampling (40  CFR Part 50, Appendix L, Section 7.4.5.2). If the sampler's flow measurement
 system is properly calibrated, Va should be accurate, and no further sample volume calculations
 are required.
Table 12-1.   Formulas for Calculations Associated with PM2.s Monitoring

  Calculation                               Formula           Equation No.     Section

  Alternative calculation of total             Va = Qavg x t x 1 0~3           12-1          12.2.1
  sample volume, used only if Va is not
  available directly from the sampler

  Determination of PIVhs mass            fl/fc.s = (Mr - Mi) x 103          12-2          12.2.2
  collected on filter (net weight)
  Determination of PIVh s mass              PM25 = M25/Va            12-3          12.2.3
  concentration
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
        PM2.5 Monitoring Data                                   January 201 6     Page 2 of 7

 NOTE: In the event the total sample volume measurement from the sampler is not available, the
 total sample volume may be calculated by multiplying the average flow rate, in actual m3/min,
 by the elapsed sample collection time in minutes. Both of these measurements are required to be
 provided by reference and equivalent method samplers. Use the following formula only if Va is
 not available directly from the sampler:
                                            Xt Xl0-3                              (12-1)
 where

          Va  = total sample volume, actual m3
        Qavg  = average sample flow rate over the sample collection period, L/min
           t  = total elapsed sample collection time, min
        10"3  = units conversion (m3/L).

 For example, a sampler with an average flow rate of 16.7 L/min (Qavg) for a 1,410-min (23.5-
 hour) sampling period (i) would have a total sample volume (Va) of 23.5 m3.

 12.2.2   Net PM2.s Mass Calculation

 The mass of parti culate matter collected on the filter during the sampling period is determined
 by subtracting the initial (tare) mass of each filter from the final mass of the filter, as

                                 M2.s = (Mf- Mr) xlO3                             (12-2)
 where

        M.2.5  = total mass of PIVfo.s collected during the sampling period, ug
         Mf = final mass of the equilibrated filter after sample collection, mg
         Mi  = initial (tare) mass of the equilibrated filter before sample collection, mg
         103  = units conversion (ug/mg).

 For example, a filter that weighed 139.293 mg before sampling (Mi) and 139.727 mg after
 sampling (Mf) would have a PM2.5 mass (M2.s) of 434 ug.

 12.2.3   PM2.5 Concentration Calculation

 Each PM2.5 mass concentration measurement is calculated by dividing the total mass of PM2.5
 (Equation 11-2) collected during the sampling period (M2.s) by the total volume of air sampled
 (Va) (taken directly from the sampler readout display or calculated from Equation 12-1), as

                                    PM2.5=M2.5/Va                               (12-3)

 For example, a sample with a mass (M2.s) of 434 ug collected from a total sample volume (Va)
 of 23.5 m3 corresponds to a PM2.5 concentration (PM2.s) of 18.5 ug/m3.
                                              2.12-PM2.5 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
        PM2.5 Monitoring Data                                    January 2016     Page 3 of 7

 12.3   Verification of Manual Calculations and Data Entry

 This section applies to calculations performed using manually recorded or transcribed data and
 when hand calculators, computerized spreadsheets, databases, or other calculation aids are used
 by field operators or weighing technicians. This section also describes verification procedures
 for data keyed from handwritten forms into a computer.

 12.3.1   Verification of Manual Calculations

 Calculations should be verified when new sites are set up, when new personnel are trained, and
 when calculation aids such as spreadsheets are changed or updated. These calculations should
 be reverified periodically as part of the regular audit program. The basic procedure for verifying
 manual calculations is as follows:

    •   Gather the raw data sources to be checked. These might include the analyst's notebook or
       the field operator's report sheets on which the data were originally recorded.

    •   Obtain a copy of the resulting data report. The report should represent the data after they
       have been entered into the monitoring organization's electronic data system. This will
       ensure that all sources of human error,  including keying errors, are included in the check.

    •   Independently verify the results based on the raw data. If calculation aids were used, do
       not use the same spreadsheet or calculator program that was originally used; the
       verification calculations  should be done as independently as possible.

    •   Verify that the correct formulas, conversion constants, and reporting units were used.

 The amount of data to be recalculated and verified depends on several factors. If the original
 calculations are suspect (for example, where mistakes have been found in a certain individual's
 work), all the questioned results should be recalculated. When there is no reason to suspect
 widespread errors, as with recalculations done as part of a routine audit, only a percentage of the
 data set need be checked initially. A commonly used audit guideline is to check 7 percent of
 manual calculations, provided that at least one example of each type of calculation is checked.
 This is a flexible guideline that depends on the amount of data to be checked and other factors.
 A cross-section of data should be selected for checking (e.g., the first and last values in a data
 set, extreme values, atypical values). The monitoring organization's QAPP should describe the
 frequency of these verifications, procedures to be followed, acceptance criteria, and corrective
 actions.

 12.3.2  Verification of Manual Data Entry

 Two methods are typically used for keystroke data entry quality control (QC): duplicate keying
 and proofing. Duplicate  keying is defined as data entry by two different operators, followed by
 resolution of all discrepancies by referring back to the original forms. Proofing refers to manual
 (visual) comparison of data entered by a single operator against the original forms. The person
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
         PM2.5 Monitoring Data                                    January 2016     Page 4 of 7

 who proofs the data should not be the person who entered it originally. Duplicate keying is far
 more likely to  eliminate simple keystroke mistakes than proofing. It is also more cost-effective
 for large data sets but requires somewhat more work to set up initially.

 If a data set has not been entered by the duplicate keying technique, a certain percentage of
 residual errors is to be expected. Keystroke data entry error rates can be estimated by comparing
 the original report pages or logbooks to the data reported out of the data system. The end use of
 the data should be considered in determining whether the error rate is unacceptably high. For
 example, if the data will be used for purposes of compliance or litigation, any avoidable errors
 may be considered unacceptable. In such a case, the data should be rekeyed as described above.

 If a data set has been entered by an effective duplicate keying procedure, simple keystroke
 errors are expected to be virtually nonexistent (less than 1 in 1,000,000, even with key operators
 of modest ability). If a check of duplicate-keyed data turns up a significant error rate, the
 integrity of the duplicate keying procedure should be investigated by examining the SOP for
 deficiencies. If no procedural problems are found, noncompliance with the SOP or hardware
 problems should be investigated.

 In addition to typographical errors, other potential sources of data entry errors should also be
 considered while performing an audit of data entry. These include poor handwriting on the
 original data forms and smears or water damage that renders the writing difficult to read.
 Hardcopy records must be maintained for a period of time so that audits and verification checks
 can be performed. The monitoring organization's QAPP should describe the policies regarding
 verification of manual data entry, audit frequency, and retention policies for original records
 needed to verify the electronic data.

 12.4   Validation

 12.4.1   Data Validation

 Validation of monitoring data is an extensive topic that can only be touched on in this section.
 The general principles are presented in Volume I of the Quality Assurance Handbook.
 Validation data screening is done for two primary reasons:

   •   To verify that the data have been recorded, entered, and calculated correctly—Validation
       screening can sometimes be used to identify problems such as failing equipment, siting
       problems, or operator errors.

   •   To screen for potential outliers (data points that are exceptional in concentration value or
       in some other way)—Data points identified as potential outliers are not necessarily
       invalid;  validation screening simply identifies data points for further investigation.

 Data validation can be done in several different ways. A combination of techniques is often the
 most effective. The exact criteria to be applied and corrective actions taken should be described
 in the monitoring organization's QAPP.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
         PM2.5 Monitoring Data                                    January 2016     Page 5 of 7
 Some techniques that may be applicable to PIVb.s data validation are the following:

    •   Graphing and visually examining time series of operating parameter data such as flow
       checks (e.g., control charting)

    •   Graphing and visually examining scatter plots of data (e.g., collocated sample results)

    •   Range checking (does the data fall within an expected range of values?)

    •   Statistical checking (flagging data points outside 3-sigma or other statistically derived
       limits)

    •   Evaluation of goodness of fit and linearity (applicable to linear regression data such as
       calibrations)

    •   Regular review of operators' notes and communication with operators to identify
       problems

    •   Review of chain-of-custody

    •   Review of audit results to identify data potentially impacted by audit findings

    •   Adherence to the monitoring organization's QAPP and SOPs.

 The appropriate decision maker should take into consideration all QC and QA data when
 making validity decisions.

 With regard to data identified as outliers: Being flagged as a potential outlier by the methods
 listed above is not prima facie evidence that a data point is invalid. Outliers must be thoroughly
 investigated before they can be excluded. Outliers  must be presumed to be correct unless there
 is documented evidence of an equipment malfunction or other exceptional condition or event
 that renders the result unrepresentative. This is particularly important because the 24-hour
 compliance criterion is based on the upper  98th percentile point of the concentration
 distribution. Because operational problems can sometimes cause anomalously high readings,
 potential outliers in the upper tail of the concentration distribution should be validated  carefully.
 Special flags are available in AQS for indicating when data are suspect.

However, all data - not just outliers - should be rigorously reviewed during the data validation
process. The monitoring organization should provide a detailed description of its data screening
procedures in its QAPP and/or in an approved data validation SOP. Decision rules for data
invalidation and for taking corrective actions should be clearly described. Corresponding AQS
data flags should be specified in the monitoring organization's written procedure.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
        PM2.5 Monitoring Data                                    January 2016    Page 6 of 7

 12.4.2 Validation of Software Used to Process PMi.s Data

 Software used to process, manage, and report PM2.5 data used for compliance purposes should
 be validated to ensure that it is free of incorrectly coded calculations and errors. The process of
 validating software is distinct from data validation discussed in the previous section.

 A structured approach to software development, testing, and validation is strongly
 recommended for managing PM2.5 compliance data. Information on structured  software
 development, testing and validation can be found in EPA QA/R-5, EPA Requirements for
 Quality Assurance Project Plans for Environmental Data Operations, Appendix G, and in many
 other publications.

 A few of the areas that should be tested during the development of PIVb.s compliance and
 reporting software include the following:

   •   Correctness of calculations;

   •   Correct assignment of input and output values (i.e., verify that the input and output values
       are correctly identified as to station, sample time period, sampler number, filter number,
       and so on);

   •   Correct calculation of statistics, including determination of the correct 98th percentile
       point for the 24-hour compliance criterion (40 CFR Part 50, Appendix N); and,

   •   Correct application of validation procedures, range checks, statistical control limits, and
       so on.

 Use of spreadsheets for processing and managing large environmental data sets is strongly
 discouraged. Spreadsheets are difficult to test thoroughly and can develop new problems as
 more data are added and predefined data ranges overflow. Modern relational database products
 offer a much more reliable environment for large-scale data processing, although they are
 somewhat more difficult to learn at first.

 Although software does not usually "break" or wear out after it has been written and tested, any
 change in its runtime environment may potentially cause problems. Therefore, periodic
 reverification of software is advisable, especially after significant changes to system hardware
 or software. Whenever individual software modules are changed, run test procedures to detect
 any unintended consequences of the changes. Software documentation that describes
 configuration changes and test results should be maintained.

 All monitoring agencies are strongly encouraged to develop and comply with SOPs for software
 development, testing, maintenance, configuration control, documentation, and data archiving.
 The QAPP should identify the relevant data processing SOPs, and these should be addressed
 during periodic audits.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.12.0 Calculations, Validations, and Reporting of
        PM2.5 Monitoring Data                                    January 2016    Page 7 of 7

 12.5   Data Reporting

 The primary standards for particulate matter in ambient air are based on the measured mass
 concentration of PM2.5. Information on reporting and interpretation of PIVfo.s data with respect to
 the attainment of these standards is covered in 40 CFR Part 50, Appendix N.

 12.5.1   Rounding

 Raw data from the samplers are preferably recorded electronically and/or communicated by
 electronic means to a central computer where they are recorded. Electronically stored data and
 computerized calculations should not be rounded until  final formatting for reports or for
 delivery to AQS.

 Filter weight data for calculating PIVfo.s concentrations  should be recorded to the full readable
 precision of the microbalance. Hand-calculated weight differences should be carried to the same
 number of digits and should not be rounded.

 For other hand-recorded data, including calibration and audit data, it is advisable to read and
 record somewhat more resolution than will be needed for calculating the final result. For
 example, to make a measurement that will be reported to the nearest 1 percent, attempt to read
 the data to at least the nearest 0.5  percent or better. It is usually not necessary to record a
 resolution more than a factor of 10 better than that needed in the final result, however.

 Special care should be taken when recording data that will be used in a difference calculation:
 enough significant figures must be recorded so that the result will have the necessary
 resolution. The most obvious example is the determination of PIVfo.s mass by the difference of
 two filter weights. Measurement of the filter weight to  six significant figures is necessary in
 order to get adequate precision in the net particulate mass. All  data recording forms should
 specify the appropriate number of digits, or the necessary resolution in physical units, for each
 parameter that must be manually recorded.

 12.5.2   Rounding Rules for Comparison with  the National Ambient Air Quality
         Standards

 For comparing calculated concentration averages to National Ambient Air Quality Standards
 (NAAQS) levels, it is necessary to use round-off rules defined in Sections 4.3 of 40 CFR Part
 50, Appendix N. For the annual PM2.5 standard (currently  12 ug/m3), the 3-year average annual
 means is rounded to the nearest tenth of a ug /m3. Decimals x.x5 and greater are rounded up to
 the next 0.1, and any decimal lower than x.x5 is rounded down to the nearest tenth of a ug/m3.
 For the 24-hour PM2.5 standard (currently 35 ug/m3), the 3-year average of the annual 98th
 percentile values is rounded to the nearest 1 / ug/m3. Decimals 0.5 and greater are rounded up to
 the nearest whole number, and any decimal lower than 0.5 is rounded down to the nearest whole
 number.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management                          January 2016     Page 1 of 6


 13.0  Data and Records Management

 13.1   Overview

 It is important to keep good records in any air pollution measurement program. This is
 particularly true for measurements of PM2.5 by reference or Class I equivalent methods because
 these are manual methods that involve the transfer and handling of samples and data by several
 people. Automated methods, on the other hand, present the final, averaged data with much less
 human intervention and, thus, limit the possibilities of transcription errors and misplacement of
 data.

 Section 14 of the U.S. EPA's Quality Assurance Handbook for Air Pollution Measurement
 Systems, Volume //provides guidance on data acquisition and information management.
 Although it emphasizes automated data acquisition, much of the discussion is also applicable to
 manual methods. There are several reasons for maintaining complete, orderly records and
 properly managing data. Records can:

   •   Provide information on mechanical problems that occur and document how the problems
       were corrected;

   •   Provide a history of warranty repairs;

   •   Provide a history of in-house repairs and preventive maintenance servicing;

   •   Document date and site placement details for the primary and collocated samplers, as
       well as the characteristics of the surrounding land areas, sources, and other features;

   •   Be a useful source of information at the time of the annual network review and/or a
       technical systems audit to show proper sampler installation  and operation, performance  of
       quality control (QC) and quality assurance (QA) checks, traceability of equipment and
       standards, and proof that all systems were kept in control (use of control charts); and

   •   Provide defensible evidence to support the quality of PIVb.s  data submitted to the national
       database (AQS).

 13.2   Methodology for  Data and Records Management

 A systematic approach to maintaining records and managing data for  all reported ambient air
 pollutant concentrations usually involves the following considerations:

    •   Personnel—One or more persons should be identified within  the monitoring
        organization to be responsible for maintaining the records and preparing and submitting
        the required data products. Education, training, and experience are requirements for the
        position. Standard Operating Procedures (SOPs) will also be required.

    •   Quality assurance—An individual or group should be assigned to prepare SOPs, to


                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management                          January 2016     Page 2 of 6

        conduct internal systems and performance audits, and to review & validate the data
        before submission.

    •   Facilities and equipment—A designated storage location for paper records and
        computer data files should be established (i.e., Local Area Network, etc).  Using file
        cabinets that can be locked is suggested. A backup copy of data files should be kept in
        another location for safety. Computers and data entry stations, as well as connections to
        centralized databases (such as AQS), are also required.

 13.3   PM2.5 Records to Create  and Retain

 A quality assurance project plan (QAPP) for the various aspects of the PM2.5 sampling program
 must be prepared. SOPs are developed as part of the QAPP. Much of the required information
 can be drawn from that given in the applicable sections of the Code of Federal Regulations
 (CFR), the sampler's instruction manual, and this document. Records to maintain are
 summarized below.

 13.3.1   Sampler Siting and Maintenance Records

 Documentation regarding  siting and maintenance of the PIVfo.s samplers should be retained in
 the files. These documents may include the following:

    •   Site selection criteria checklist; photographs of the site in  the eight cardinal directions,
        with the PM2.5 sampler in the center foreground; site sketch; U.S. Geological Survey
        (USGS) topographical map of the area; and a copy of the  site documentation records that
        were submitted to the AQS database when the site was first established and any updates
        made since that time.

    •   Siting documentation in annual network plans (ANP), which are posted on AMTIC.

    •   Procurement log for field equipment (see Figure 4.1). Notes on acceptance/rejection
        tests (see Table 4-1).

    •   Warranty and maintenance records for each PIVfo.s sampler. A dated maintenance
        checklist should be established for each sampler; separate maintenance logbooks should
        be used for the PIVfo.s samplers.

    •   Manufacturer-supplied calibration and traceability records for the thermometers, flow
        rate measuring devices, and pressure sensors used for calibrating, checking, or auditing
        PM2.5 samplers.

    •   Chain-of-custody records.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management                           January 2016     Page 3 of 6

 13.3.2   Analytical Laboratory Installation Records

 Records to keep about the physical setup of the gravimetric laboratory include the following:

   •  Equipment inventory and warranty information (microbalances, anti-static devices,
      calibration and check weights, and so on);

   •  Procurement log (see Figure 4.1), and notes on acceptance/rejection tests (see Table 4-1);

   •  Records (e.g., control charts) that demonstrate that the temperature and relative humidity
      (RH) within the weighing laboratory were held within specified limits;

   •  Minute data logs or spreadsheets containing the raw temperature and relative humidity
      data from sensors;

   •  Microbalance logbook; and,

   •  Certification records (NIST-traceability) for all standards in use in the lab.


 13.3.3   Field Sampling Operation Records

 PM2.5 sampling operations generate a number of paper and/or computerized records that need to
 be reviewed, reported, and filed. Starting with the calibration of the sampler's sensors, these
 records will include at least the following items:

   •  Sampler calibration logbooks or data sheets (see Figure 7.3 for an example) for entering
      results of temperature, pressure, and flow rate checks, audits, and calibrations. Also
      include documentation of the devices used to calibrate the sampler(s) and these devices'
      traceability records or references to the traceability records. A separate  section of the
      logbook should be used for each sampler in the network. Enter results of the sampling
      procedure checks (see Table 7-1), which include notes on filter inspection, and checks of
      flow rate and leaks.

   •  PM2.5 sampler field data sheet (see Figure 7.1  for an example). The run data sheet should
      be prepared at least in duplicate by using carbonless paper or making copies. The original
      of the data sheet should accompany the filter sample to the laboratory, and a copy should
      be retained in the field site records files.

   •  Files downloaded from the data port of each PM2.5 sampler.

 13.3.4   Weighing Laboratory Operation Records

 Data records from activities in the weighing laboratory are expected to include at least the
 following items:
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management                          January 2016    Page 4 of 6

   •   Records of temperature and humidity control in the filter conditioning environment. Use
       of control charts is recommended.

   •   Laboratory data forms (see Figure 10.5 for an example).

   •   Laboratory database files (if utilized).

   •   Results of microbalance calibrations and servicing. This information can be recorded in
       the microbalance logbook.

   •   Results of filter integrity checks and determinations of the conditioning periods (i.e., lot
       stability testing) required for various filter batches (refer to Sections 10.5,  10.6 and 10.7).

   •   The completed PIVb.s sampler run data sheet and/or COC (see Figures 7.1  and 7.2 for
       examples).

   •   Records of sample numbers (identifications) and locations of archived PIVb.s samples.

 13.3.5  Quality Assurance Records

 Quality assurance systems and performance audits require complete documentation as well.
 Records of audits may be similar to the following items:

    •   Flow rate, temperature, and pressure audit data sheet (see Figure 11.1 for an example)

    •   Systems audit questionnaire (see Figure 11.3 for an example), and subsequent written
        report (if applicable)

    •   Data review/validation records.


 13.4   Data Reporting Requirements

 The EPA Regional Office and the AQS database require the following information be reported
 on a quarterly (generally, calendar-quarter) basis (EPA will calculate measurement uncertainties
 for the entire calendar year):

    •   Siting documentation (upon installation and any changes thereafter)

    •   PM2.5 concentration data (ug/m3, entered electronically into the AQS database)

    •   Information calculated and provided by the sampler (refer to Table 13-1 [Adapted from
        40 CFR Part 50, Appendix L, Table L-l]). This information is retrieved from the data
        port of each sampler. It is generally downloaded in the field to a laptop computer or
        another data storage device and  later processed in the laboratory or office. Data outputs
        to be provided to the AQS database are indicated in Table 13-1 by the bullet symbol.


                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management
January 2016     Page 5 of 6
     •   Results of all valid flowrate verifications and audits, collocated precision data, and PEP
         audits performed during the quarter. Refer to Section 11.0 for details.
 Table 13-1.  Summary of Information to be Provided by Sampler
Information to be provided
Flow rate, 30-second maximum
interval
Flow rate, average for the sample
period
Flow rate, CV, for sample period
Flow rate, 5-min average out of
spec. (FLAG5)
Sample volume, total
Temperature, ambient, 30-second
interval
Temperature, ambient, min., max..
average for the sample period
Baro. pressure, ambient, 30-second
interval
Baro. pressure, ambient, min.,
max., average for the sample
period
Filter temperature, 30-second
interval
Filter temp, differential, 30-second
interval, out of spec. (FLAG8)
Filter temp., maximum differential
from ambient date, time of
occurrence
Date and Time
Sample start and stop time settings
Sample period start time
Elapsed sample time
Elapsed sample time, oul of spec.
(FLAG6)
Power interruptions £1 min., start
time of first 10
User-entered information, such as
sampler and site identification
Appendix L
section
reference
7.4.5.1
7.4.5.2
7.4.5.2
7.4,5.2
7.4.5.2
7.4.8
7.4.8
7.4.9
7.4.9
7.4.11
7.4.11
7.4.11
7.4,12
7.4.12
7.4.12
7.4.13
7.4.13
7.4.15.5
7.4.16
Availability
Anytime1
y

*
y
X
y

y
*
y


y
y

*

,
y
End of
period2

y
y
y
y

y

y

y


y
y
y
y
y
y
Visual
display3
y

*
y
y
y
y
y
y
y
y
*
y
y
y
y
y

y
Format
Data
output4

y
y
y«
y

y*

y»

y«


y
y
y
y«
y
y«
Digital
reading5
xx.x
xx.x
XX .X
On/Off
XX.X
xx.x
XX .X
XXX
XXX
xx.x
On/Off
X.X. YY/MM/DD
HH.mm
YY/MM/DD
HH.mm
YY/MM/DD
HH.mm
YY/MM/DD
HH.mm
HH.mm
On/Off
IHH.mm,
2HH.mm, etc.
As entered
Units
L/fflin
LAnin
%

m3
•C
°c
mm Hg
mm Hg
3C

°C, Yr/Mon/Day
Hrs. min
Yr/Mon/Day Hrs.
min
Yr/Mon/Day Hrs.
min
Yr/Mon/Day Hrs.
min
Hrs. min

Hrs. min

      y  Provision of this information is required.

      * Provision of this information is optional. If information related to the entire sample period is optionally provided prior
   to the end of the sample period, the value provided should be the value calculated for the portion of the sampler period
   completed up to the time the information is provided.

      •  Indicates that this information is also required to be provided to the Air Quality System (AQS) data bank; see
   §58.16 of this chapter. For ambient temperature and barometric pressure, only the average for the sample period must be
   reoorted.
                                                     2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.13.0 Data and Records Management                                 January 2016      Page 6 of 6


       1. Information is required to be available to the operator at any time the sampler is operating, whether sampling or not.

       2. Information relates to the entire sampler period and must be provided following the end of the sample period until
   reset manually by the operator or automatically by the sampler upon the start of a new sample period.

       3. Information shall be available to the operator visually.

       4, Information is to be available as digital data at the sampler's data output port specified in section 7.4.16 of this
   appendix following the end of the sample period until reset manually by the operator or automatically by the sampler upon
   the start of a new sample period.

       5. Digital readings, both visual and data output, shail have not less than the number of significant digits and resolution
   specified.

       6. Flag warnings may be displayed to the operator by a single flag indicator or each flag may be displayed individually.
   Only a set (on) flag warning must be indicated; an off (unset) flag may be indicated by the absence of a flag warning.
   Sampler users should refer to section 10.12 of this appendix regarding the validity of samples for which the sampler
   provided an associated flag warning.
                                                          2.12-PM2.5 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty            January 2016    Page 1 of 6
        for Monitoring Data
 14.0 Assessment of Measurement Uncertainty for Monitoring  Data
 14.1  Overview

 PQAOs operating SLAMS are required to assess the measurement uncertainty of their PM2.5
 monitoring data (EPA 1997). Three different procedures are used in this assessment:

    •  Flow rate audit—The accuracy of the PIVb.s sampler's flow rate is assessed by
       performing a flow rate audit as described in Section 11.2.1. Performance of this audit is
       the responsibility of the SLT and/or the PQAO. The goal for accuracy of the sampler
       flow rate is to be within ±2 percent of the value indicated by the audit device during
       calibration, +4 percent during one-point verification checks and audits, and within ±5
       percent of the sampler's design flow rate (16.67 L/min).

    •  Bias—PM2.5 measurement bias is assessed by conducting a collocated sampler audit
       (PEP audit). The U.S. EPA Regional Offices will conduct this performance evaluation.
       The bias goal is to be between -10 percent and +10 percent (i.e., to have the site
       sampler's PM2.5 concentration value agree within ±10 percent of the audit value).

    •  Precision—PM2.5 measurement precision is assessed by collocating (i.e., locating a
       second QC sampler alongside the primary sampler used for measurement) samplers and
       reporting the results of both the measurement sampler and the collocated sampler.
       Operation of the collocated samplers is the responsibility of the SLT and/or the PQAO.
       The precision goal is to have a coefficient of variation (CV) of less than 10 percent.

 These procedures are very similar to those used in the data quality assessment (DQA) of PMio
 monitoring data. The results of these procedures are used to screen individual samplers for bias
 or excessive imprecision. EPA also uses these results to assess measurement uncertainty for
 each PQAO. The results from all  SLAMS PQAOs are used by EPA to assess the measurement
 uncertainty of data from each PM2.5 reference or equivalent method on a national basis.

 14.2  Flow Rate Audits

 Flow rate audits consist of measuring the sampler's normal operating flow rate with a flow rate
 transfer standard.  This procedure  is described in Section 11.2.1. Manual PM2.5 sampler methods
 must be audited every 6 months (i.e., every 182 days). Audits should be scheduled to avoid
 interference with regularly scheduled sampling periods. Whenever possible, these flow audits
 should be performed at randomly selected times with respect to time of day and day of week.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty             January 2016    Page 2 of 6
        for Monitoring Data
 Results of the flow rate audit to be reported to EPA are as follows:

   •  The audit (true) flow rate as determined using the transfer standard.

   •  The corresponding flow rate as measured by the sampler. This rate should be that used by
       the sampler to calculate its reported total sampling volume and average flow rate. This
       information is typically obtained from the sampler's flow rate display or from its data
       system.

 The sampler's flow rate accuracy (^4) should be within ± 4 percent of the audit value.
 Furthermore, the audit measured flow rate accuracy (An) should be within ±5 percent of the
 sampler's design inlet flow rate (16.67 L/min). The sampler's flow rate accuracies (A and AD)
 are calculated as follows:
                           A(%)  =     '  Q	^^ xlOO
                                          ^Audi t
                          AD(%) = 100 x(Qoudu - 16.67) /16.67                        (2)

 where

             A = flow rate accuracy (percent)
            AD = flow rate accuracy (percent) versus design flow rate
        Qsampier = flow rate as measured by sampler (L/min)
         QAudtt = flow rate as measured by the flow rate transfer standard (L/min)
         16.67 = design flow rate (L/min).

 14.3   Bias Assessment

 Sampler bias is assessed from the results of an FRM PEP administered through the EPA
 Regional Offices. The goal for acceptable bias is between -10 percent and +10 percent (i.e., the
 reporting primary sampler's concentration should agree to within ±10 percent of the FRM PEP
 sampler's concentration).

 The individual sampler and audit measurements must be reported to EPA, which uses them to
 calculate the single sampler bias (E) and the quarterly average bias  ( B  ) for a PQAO. PQAOs
 may also want to calculate these parameters for their quality assurance (QA) use. The single
 sampler bias (B) may be calculated as follows:
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty             January 2016    Page 3 of 6
        for Monitoring Data
                                  PM25    ,  -PM2
                         £(%) =    2'5sa^r	2
                                            2-5audit
 where

             B = bias (percent)
    PM2 5      = PM2.5 concentration as measured by sampler (ug/m3)
         Sampler
      PM       = PM2.5 concentration as measured by audit device (ug/m3).
 The quarterly average bias for a single sampler (  B )  may also be calculated as follows:
                                      —     1
                                      B  = -,11,,
 where
        B  = average quarterly bias (percent)
        Bt = bias for the ith measurement (percent)
         n = number of measurements during quarter.
 14.4   Precision

 14.4.1   Overview

 PM2.5 precision is assessed by collocating QC samplers (i.e., locating a second sampler
 alongside the primary sampler used to report a measurement). Both samplers are run identically
 and the results from both samplers are reported to EPA. The collocated QC sampler's inlet must
 be placed horizontally between 1 and 4 meters from the reporting sampler and within 1 meter of
 the same height (elevation) as the reporting sampler (Section 5.1.2). If there is a high volume
 sampler at the site, both the primary and collocated PIVb.s samplers must be separated from it by
 at least 2 meters.  The collocated QC sampler must be operated and its sample processed in
 exactly the same way as the primary reporting sampler and should be operated at least every
 sixth day on the same start time/end time schedule. It is important to always designate one
 sampler as the primary or reporting sampler and the other always as the collocated (or QC)
 sampler, even when the reporting sampler may be identical in make and model to the collocated
 sampler. The precision goal is to have a CV of less than 10 percent. Refer to 40 CFR Part 58,
 Appendix A, Section 5.5, for detailed calculation procedures. Under no circumstances should
 data from the collocated sampler be reported as coming from the reporting sampler. However,
 collocated sampler data should always be reported even when the primary sampler's data are
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty
        for Monitoring Data

 nonexistent or invalid.
January 2016    Page 4 of 6
 Concentration data must be reported for both the primary and collocated sampler, no matter
 how low the concentration. Because agreement between the measurements of collocated
 samplers may be relatively poor at low concentrations, collocated measurement pairs are used
 by EPA to calculate precision only when both PM2.5 measurement concentrations are above
 3 ug/m3.

 14.4.2   Number of Collocated Samplers Required

 To assess precision of their sampling data, SLAMS PQAOs must meet the following collocation
 requirements:

 For each distinct monitoring method designation (FRM or FEM) that a PQAO is using for a
 primary monitor, the PQAO must:

    •   Have 15 percent of the primary monitors collocated (values of 0.5  and greater round up);
        and,
    •   Have at least 1 collocated quality control monitor (if the total number of monitors is less
        than 3). The first collocated monitor must be a designated FRM monitor.

Table 14-1 demonstrates this requirement.

 In addition to these requirements, additional specifications must also be observed with respect
 to the different makes and models of samplers used as collocated samplers:

    •   A primary monitor designated as an EPA FRM shall be collocated with a quality control
        monitor having the same EPA FRM method designation.

    •   For each primary monitor designated as an EPA FEM used by the PQAO, 50 percent of
        the monitors designated for  collocation, or the first if only one collocation is necessary,
        shall be collocated with a FRM quality  control monitor and 50 percent of the monitors
        shall be collocated with a monitor having the same method designation as the FEM
        primary monitor. If an odd number of collocated monitors is required, the additional
        monitor shall be a FRM quality control monitor. An example of the distribution of
        collocated monitors for each unique FEM is provided in Table 14-1, which can also be
        found in 40 CFR Part 58, Appendix A,  Section 3.2.3.4. Table 14-2 demonstrates the
        procedure with a PQAO having an FRM and multiple FEMs, which can also be found in
        40 CFR Part 58, Appendix A, as Table A-2.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty
        for Monitoring Data
January 2016    Page 5 of 6
 Table 14-1 Distribution of Collocated Monitors for Each Unique FEM
#Primary FEMS of a unique
method designation
"1-9"
"10-16"
"17-23"
"24-29"
"30-36"
"37-43"
#Collocated
1
2
3
4
5
6
#Collocated with an
FRM
1
1
2
2
3
3
#Collocated with same
method designation
0
1
1
2
2
3
 As an example (Table 14-2), consider the minimum collocation requirements for an
 organization that has 54 primary samplers composed of 20 FRM samplers, 20 equivalent
 designation A samplers, 2 equivalent designation B samplers, and 12 equivalent designation C
 samplers. Each type of sampler should be considered separately as shown in Table  14-2 for
 FEM sampler types A, B, and C.
 Table 14-2 Summary of PM2.s Number and Type of Collocation (15% Collocation Requirement)
 Required Using an example of a PQAO that has 54 Primary Monitors (54 sites) with 1 Federal
 Reference Method Type and Three types of Approved Federal Equivalent Methods.
Primary
sampler
method
designation

FRM
FEM (A)
FEM (B)
FEM (C)
Total no. of
monitors



20
20
2
12
Total no. of
collocated



3
3
1
2
No. of
collocated with
FRM


3
2
1
1
No. of
collocated with
same method
designation as
primary
3
1
0
1
 14.4.3   Location of Collocated Samplers
 Collocated samplers should be placed at sites that have the highest PM2.5 concentrations, with
 special emphasis on sites likely to be in violation of the NAAQS. Data from other particulate
 measurement methods, such as PMio samplers, can be helpful in selecting sites for collocation.

 SLAMS PQAOs that have areas in violation of the NAAQS should place their collocated
 samplers as follows:

    •   50 percent of the collocated quality control monitors should be deployed at sites with
        annual average or daily concentrations estimated to be within ±20 percent of either the
        annual or 24-hour NAAQS, and the remainder at the PQAO's discretion.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.14.0 Assessment of Measurement Uncertainty             January 2016    Page 6 of 6
        for Monitoring Data

    •   If an organization has no sites with annual average or daily concentrations within ±20
        percent of the annual NAAQS or 24-hour NAAQS, 50 percent of the collocated quality
        control monitors should be deployed at those sites with the annual mean concentrations
        or 24-hour concentrations among the highest for all sites in the network and the
        remainder at the PQAO's discretion.
 14.4.4   Schedule for Operation of Collocated Samplers

 Collocated samples should be taken to reflect the normal operation of the reporting sampler.
 The collocated samples should be evenly distributed across the various seasons and days of the
 week. A simple system to accomplish this is to run the collocated sampler every sixth day.
 Thus, if the first week's collocated sample was run on Monday, the second week's collocated
 sample would be run on Sunday, the third week's on Saturday, and so on. Remember to start
 and stop both the collocated and reporting samplers at exactly the same time for each sample
 run.
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.10.15.0 References                                           January2016    PagelofS
 15.0  References

 ASTM. 1977. Standard test methods for measuring surface atmospheric pressure. American
 Society for Testing and Materials. Philadelphia, PA. Standard D 3631-84.

 ASTM. 1992. Standard test method for inspection and verification of thermometers. American
 Society for Testing and Materials. Philadelphia, PA. ASTM Standard E77-92.

 ASTM. 1993a. Standard practice for maintaining constant relative humidity by means of
 aqueous solutions. American Society for Testing and Materials. 1993 Annual Book of ASTM
 Standards, Vol 11.03, Designation E 104-85 (reapproved 1991), pp. 570-572.

 ASTM. 1993b. Standard specification for laboratory weights and precision mass standards.
 American Society for Testing and Materials. 1993 Annual Book of ASTM Standards, Vol
 14.02, Designation E 617-91, pp. 280-295.

 ASTM. 1995a. Standard specification for ASTM thermometers. American Society for Testing
 and Materials. Philadelphia, PA. ASTM Standard El-95.

 ASTM. 1995b. Standard test methods for measuring surface atmospheric pressure. American
 Society for Testing and Materials. Publication number ASTM D3631-95.

 ASTM. 1995c. Standard practice for evaluation of air assay media by the monodisperse DOP
 (Dioctylphthalate) smoke test. American Society for Testing and Materials. Philadelphia, PA.
 ASTM Standard D2986-95a.

 Desert Research Institute. 1994. DRI Standard Operating Procedure. Gravimetric Analysis, DRI
 SOP No. 2-102.3, Reno, NV. 24 pp.

 Eisner AD. 1997. Personal communication. ManTech Environmental Technology, Inc.,
 Research Triangle Park, NC.

 Engelbrecht DR,  Cahill TA, Feeney PJ.  1980. Electrostatic effects on gravimetric analysis of
 membrane filters. Journal of the Air Pollution Control Association 30(4):319-392.

 EPA 1997b. Guidance for network design and optimum site exposure for PM2.5 and PMio. U.S.
 Environmental Protection Agency, Office of Air Quality Planning and Standards. Finalized
 version, December 1997.

 EPA 2013. Quality Assurance Handbook for Air Pollution Measurement Systems. Volume II,
 Part I. Ambient Air Specific Methods. Appendix 12, "Calibration of Primary and Secondary
 Standards for Flow Measurements."

 EPA.  1994. EPA requirements for quality assurance project plans for environmental data
 operations. U.S. Environmental Protection Agency. Document EPA QA/R-5. External Review
 draft final, October 1998.

                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.10.15.0 References                                           January 2016    Page 2 of 3
 EPA. 1995. Guidance for the preparation of standard operating procedures (SOPs). U.S.
 Environmental Protection Agency. Publication QA/G-6, U.S. EPA Quality Assurance Division,
 Washington, DC. November.

 EPA. 1997. Ambient air quality surveillance. U.S. Environmental Protection Agency. 40 CFR
 Part 58, Appendix A—Quality assurance requirements for State and local air monitoring
 stations (SLAMS); Appendix E—Probe and monitoring path siting criteria for ambient air
 quality monitoring, as revised July 18, 1997.

 EPA. 1997. Reference method for the determination of fine particulate matter as PIVh.s in the
 atmosphere. U.S. Environmental Protection Agency.  40  CFR Part 58, Appendix L, as amended
 July 18, 1997.

 EPA. 1997. Reference method for the determination of fine particulate matter as PIVb.s in the
 atmosphere. U.S. Environmental Protection Agency.  40  CFR Part 50, Appendix L.

 EPA. 1997a. Network design for State and local air monitoring stations (SLAMS), national air
 monitoring stations (NAMS), and photochemical assessment monitoring stations (PAMS). U.S.
 Environmental Protection Agency. 40 CFR Part 58, Appendix D, as amended July 18, 1997.

 EPA. 1997a. Reference method for the determination of fine particulate matter as PM2.5 in the
 atmosphere. U.S. Environmental Protection Agency.  40  CFR Part 50, Appendix L.

 EPA. 1997b. Ambient air monitoring reference and equivalent methods. U.S. Environmental
 Protection Agency. 40 CFR Part 53, as amended July 18, 1997.

 EPA. 1997c. Ambient air quality monitoring methodology. U.S. Environmental Protection
 Agency. 40 CFR Part 58, Appendix C, as amended July  18, 1997.

 EPA. 1997c. Reference method for the determination of fine particulate matter as PM2.5 in the
 atmosphere. U.S. Environmental Protection Agency.  40  CFR Part 50, Appendix L.

 EPA. 1997d. Ambient air monitoring reference and equivalent methods. U.S. Environmental
 Protection Agency. 40 CFR Part 53, as amended July 18, 1997.

 EPA. 2008. Quality Assurance Handbook for Air Pollution Measurement Systems Volume IV:
 Meteorological Measurements Version 2. U.S. Environmental Protection Agency. Document
 No. EPA-454/B-08-002. Revised March.

 Harris G.  1993. Ensuring accuracy and traceability of weighing instruments. ASTM
 Standardization News 21(4):44-51.

 Harris GL. 1994. State weights and measures laboratories: state standards program description.
 National Institute of Standards and Technology. Special publication 791. 130 pp.

 Hawley RE, Williams CJ. 1974. Electrostatic effects  in microbalances. I. General considerations

                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.10.15.0 References                                            January 2016    Page 3 of 3

 of the effectiveness of a radioactive ionizing source under ambient conditions. Journal of
 Vacuum Science and Technology ll(l):419-422.

 Kupper WE. 1990. High accuracy mass measurements, from micrograms to tons. Instrument
 Society of America Transactions 29(4): 11-20.

 Lipfert FW. 1994. Filter artifacts associated with particulate measurements: recent evidence and
 effects on statistical relationships. Atmos Environ 28:391-392.

 Nelson, G.O. 1992. Gas Mixtures: Preparation and Control. Lewis Publishers, Boca Raton, FL,
 304 pp.

 NIST. 1976. Liquid-in-glass thermometry. National Institute of Standards and Technology.
 NBS Monograph 150. January.

 NIST. 1986. Thermometer calibration: A model for State calibration laboratories. National
 Institute of Standards and  Technology. NBS monograph 174. January.

 NIST. 1988. Liquid-in-glass thermometer calibration service. National Institute of Standards
 and Technology. Special publication 250-23. September.

 NIST. 1989. The calibration of thermocouples and thermocouple materials. National Institute of
 Standards and Technology. Special publication 250-35. April.

 Papp ML, Elkins JB, Musick DR, Messner MJ. 1998. Data Quality Objectives for the PM2.5
 Monitoring Data. U.S. Environmental Protection Agency, Research Triangle Park, NC. In
 preparation.

 Weil J. 1991. Static control for balances. Cahn Technical Note. Published by Cahn Instruments,
 Inc., Madison, WI.

 White VR. 1997. National Voluntary Laboratory Accreditation Program: 1997 Directory.
 National Institute of Standards and Technology. Special publication 810. 225 pp.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                  January  2016     Page 1 of 12

 Appendix A—Measuring Alkalinity of Filters


 Note to the reader. The procedure given here is one created for and extensively tested for use
 with glass fiber or quartz high-volume filters. It is presented as guidance to the analyst for
 devising and testing a similar procedure for determining the acid-base character ofPM2.s
 sampling filters.

 A.1    History

 Alkalinity of hi-vol filters is measured because alkaline sites react with SCh, NOX and other acid
 gases in the air forming sulfates, nitrates, etc.  Such compounds' masses are indistinguishable
 from particulate mass collected by the filter. In turn, the total particulate weight, sulfate,
 content, nitrate content, etc., would be falsely high in proportion to filter alkalinity.

 A.2.   Scope and Applicability

 This method describes procedures for measuring the alkalinities of glass and quartz fiber filters.
 New filters are tested for compliance with EPA alkalinity specifications and older filters are
 tested in order  to evaluate appropriate correction factors for previously obtained particulate data.
 Table A-l outlines the procedure.

 A.3    Equipment and Reagents

 Table A-2 lists the equipment and reagents. The supplier is always Fisher Scientific Company,
 unless otherwise specified. (This information  is for reference purposes only.)

  TABLE A-1.  GENERAL DESCRIPTION OF EPA FILTER ALKALINITY PROCEDURES

          Parameter                              EPA specification
  Sample container            250-mL Phillips  beaker
  hhO extraction volume, ml    100

  Heating technique           Hotplate, reflux condenser (sample temperature about 100 °C)
  Extraction time, min           20
  Mixer                      Magnetic stirrer with 2-in stirring bar
  Filtration equipment           7-cm ID unheated Buchner funnel 250-mL filter flask
  Filtration procedure           Twice through Whatman 41 or 541
  Titration vessel              250-mL filter flask (no transfer needed)
  Titration technique           Potentiometric without temperature control or exclusion of atm  CO2
  Endpoint                    pH = 7.00 ± 0.01 after 5 min equilibration (electrode stationary)
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                  January 2016    Page 2 of 12




                      TABLE A-2. EQUIPMENT AND REAGENTS
Minimum Qty
Equipment
1 roll
1 pair
1 pair
1
2 pairs
1
100
1 pair
1
2
2
1
1
2
1
3
2 (optional)
3
1
1
1
1
1
1
2
3
6
1 or2
1 or2
1
2
1
1
1
Description

Heavy duty aluminum foil or stainless steel surface
Scissors or shears
Forceps, 300 mm in length
Ruler, 12 in. (300 mm) hardwood or metal
Lint-free gloves
Analytical balance (accurate to 1 mg)
Aluminum weighing dishes
Heat-resistant gloves
pH Meter (accurate to 0.01 pH units)
Combination pH electrode with AgCI reference and ceramic
junction
Electrode storage containers
Rectangular porcelain stand
Double buret clamp for above
Buret, 10 ml graduated in 0.05 ml, Class A
Buret, 25 ml or 50 ml graduates in 0.1 ml, Class A, 3-way
stopcock optional
Beakers, 10 ml or buret covers
Aspirator bottle with tubulation, 1 ,000 ml (storage reservoir
for H2SCM titrant)
Volumetric flasks, 1 ,000 ml, Class A
Volumetric flask, 500 ml Class A
Pipet, 1 .00 ml, Class A
Pipet, 4.00 ml, Class A
Pipet, 10.00 ml, Class A
Pipet, 100.00 ml, Class A
Extension-type ring, 2-3 in. or clamp
3-prong clamp, 57 mm grip size
3-prong clamp, 89 mm grip size
Clamp holders
Ring stands
Lead bricks for above (or similar weighted object)
Hotplate/stirrer, Corning PC-351 or equivalent
Hotplates, Corning PC-35 or equivalent
Watch glass, 90-mm diameter
Erlenmeyer flask, 2,000 ml
Neoprene or rubber stoppers #6 1/4, with hole to fit Buchner
funnel
Supplier, Cat. No.

01-213-11
14-275C
10-316C
12-090
11-393-5

08-732
Lab Safety Supply
H 1955-2


13-640-300
14-667
05-779
03-700-22A
03-724-1 OA
02-539C
02-972D
10-210G
10-21 OF
13-650-2B
13-650-2E
13-650-2L
13-650-2U
14-055B
05-740
05-742
05-757 or 05-759


11-495-50
11-495
02-61 2C
10-040M
14-141H
                                           2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters
January 2016    Page 3 of 12
Minimum Qty
2 (optional)
1
1
1
As needed

1 (2nd optional)
1
1
1
1
1
1
2
2
1
9
10
Various
1
1
1
1
As needed
1
1
1

Reagents
1
1
1
Description
#8, for aspirator bottle
#7, with hole to fit condenser
#7, with hole to fit tube below
1/4 in. to 3/8 in. O.D. heavy-walled glass or metal tube
approximately 5 to 6 in long
Vacuum tubing
Rubber
Tygon
Tubing connector
Y-type
T-type
Vacuum source (house vacuum, water aspirator, etc.)
Buchner funnel for 7 cm filter paper
Box of filter paper, Whatman 41 (or 541), 7 cm
Forceps, 150 mm L
Stopwatch or equivalent timer, accurate to 1 sec or better
Spatula
Wash bottles, polyethylene
Tall form beakers, 200 ml or 150 ml beakers
Graduated cylinder, 100 ml
250 ml Phillips beakers or 250 ml wide mouth flasks
Filter flasks, 250 ml
Assorted stir bars Teflon® coated or glass beads
Magnetic stir bar >2 in length, Teflon coated
Beaker, 400 ml
Water-cooled condenser, must be fitted to Phillips beakers or
wide mouth flasks above
Filter flask 500 ml
Laboratory tissues and towels, sizes as available
Box glassine envelopes, #7 4 1/8 in. x 6 1/4 in., ungummed
flap
Face shield
Box hose clamps


Case pH standard buffer solution, pH 7.00
Case pH standard buffer solution, pH 10.00
0.0500 N, Sodium Hydroxide NaOH, Fisher CERTIFIED,
Supplier, Cat. No.
14-141K
14-1 SOL
14-1 SOL

14-173C
14-169-2B
1 5-320-1 OC
15-319C

1 0-356C
09-850A
1 0-275
14-648 or 14-649-5

03-409-1 OE or 03-
409-22C
02-545B
08-549-5E
02-566B
10-1 81 D
10-1 81 D
14-511-65
02-540L
07-723B
10-180E

Worcester
Envelope Co.,
#3525
11-409-5
American Scientific
Products, C6096-
232

SB1 07-500
SB1 15-500
SS278-1
                   C.A.S. Reg. 1310-73-2, 1-L bottle
                   1.000 N, Sulfuric Acid, H2SO4, Fisher CERTIFIED, C.A.S.
                   Reg. 7664-93-9, 1-L bottle
          SA212-1
                                                2.12-PM25 Reference/Class I  Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                   January 2016    Page 4 of 12

   Minimum Qty    Description                                            Supplier, Cat. No.
         1          Potassium Chloride, CKI, Fisher CERTIFIED, Crystals,         P-217-500
                   C.A.S. Reg. No. 7447-40-7, 500-g bottle
         1          Hydrochloric Acid, HCI, ACS reagent, C.A.S. Reg. No.         A-144-212
                   7647-01-01/2 gal bottle
                   Deionized water                                         Laboratory
                                                                         Deionization
                                                                         System
 A.4    Procedure

 A.4.1    Sample Preparation

 Filters should be handled only with clean lint-free gloves or clean metal utensils such as
 forceps, tongs, scissors, etc.

 (1)  After initial receipt, place the Low-vol filters in a metal rack with separate slots for each
     filter. Place this rack in a controlled humidity (<50 percent relative humidity), controlled
     temperature (25 °C), balance room and allow the filters to equilibrate for 24-48 hours. At
     the end of that period, carefully weigh each filter using a balance adapted for this purpose.
     Weigh to the nearest milligram or better. Then place the filters in a storage container.

 (2)  Remove one filter at a time from the storage container.

 (3)  Place the filter on a clean metal surface (aluminum foil spread on a regular lab bench, clean
     stainless steel surface, etc.). Measure its length and width on all edges. Calculate the area
     by multiplying the average length by the average width. Cut out the filter number printed in
     the upper right corner, in a single rectangular piece. Save the number and place it in a
     glassine envelope.

 (4)  Using a ruler, scissors, and lint-free gloves, cut the filter into /^ in. squares. This makes
     approximately 320 squares.

 (5)  Mix the squares and randomly select a sample of 70-80 squares (approximately 1 g). Do
     not select and analyze the piece of filter containing the filter number. Weigh the filter
     sample in a tared disposable aluminum weighing dish. The final sample weight must be
     within one square of 1 g. Record the weight to the nearest milligram.

 (6)  Store the remaining squares in a glassine envelope with the filter number showing. NOTE:
     Filters may be cut a few days before analysis and stored in glassine envelopes.

 (7)  Transfer the weighed sample into a 250-mL Phillips beaker or a 250-mL wide-mouth flask.
     Cover the beaker or flask with the aluminum weighing dish and seal.

 (8)  Clean the gloves,  utensils and work area of any filter fibers or lint before taking out the
     next filter.
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                   January 2016    Page 5 of 12
 (9)  Repeat Steps 2-8 until seven Phillips beakers containing filter squares have been prepared.
     Leave two Phillips beakers empty as blanks. Label the Phillips beakers and 250-mL filter
     flasks consecutively from 1  to 9 and pair by number for each sample or blank.

 A.4.2   Solution Preparation

 (1)  Prepare 0.005 N NaOH titrant by diluting a certified 0.0500 N solution 1:10 with degassed
     deionized water (i.e., pipet 100.0 mL of certified NaOH into a 1,000-mL volumetric flask,
     dilute to the mark, and mix well). Degas the deionized water by bubbling N2 through 1 L of
     water at a rate of approximately 1 L/min for 1 hour. Take care to protect the solution from
     the air during storage in order to prevent carbonate formation.

 (2)  Prepare 0.0100  N H2SO4 titrant by diluting certified 1.000 N HXSO4 titrant by diluting
     certified 1.000 NIHhSCU solution 1:100 with deionized water (i.e., pipet 10.00 mL into a
     1,000-mL volumetric flask,  dilute to the mark with deionized water, and mix well).

 (3)  Prepare 0.1 M KC1/0.0001 M HC1 electrode storage solution by the following process.
     First, pipet 1 mL of reagent  grade concentrated HC1 into a 500-mL volumetric flask, dilute
     to the mark with deionized water, and mix well. Second, place 7.456 g of KC1 in a 1,000-
     mL volumetric flask, add approximately 500 mL deionized water, swirl to dissolve the KC1
     crystals, pipet 4 mL of the diluted HC1 solution into the flask, dilute to the mark with
     deionized water, and mix well. The solution may be stored in the volumetric flask or
     similar glass container.

 A.4.3   General Notes and Pre-run Setup

 (1)  Set up the titration stand and burets well in advance of sample analysis. Place one buret
     clamp on the titration stand  and adjust the  clamp height. In the right-hand buret holder,
     place a 10-mL buret. Check to see that the stopcock is closed. Fill the buret with 0.005 N
     NaOH solution  and place a small beaker or buret cover on top of the buret:
                                       CAUTION

           The NaOH solution should not be stored in the buret for any
           length of time due to its tendency to form carbonates when
           exposed to the air.
     Place the other buret in the other side of the holder, fill the buret with 0.01 N H2SO4, and
     cover the top of the buret with a small beaker or buret cover.

 (2)  (Optional: If using buret with 3-way stopcock.) Attach one end of a length of acid-resistant,
     flexible tubing to the curved tube, A, at the bottom of the buret and connect the other end to
     a 3-way tubing connector. Securely fasten both ends using hose clamps. Attach the other
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                   January 2016    Page 6 of 12

     two branches of the tubing connector with flexible tubing to the tubes at the base of the
     aspirator (reservoir) bottles. If a single reservoir bottle is used, curved tube A may be
     attached directly by a single line. Again, secure the connections with hose clamps. Check
     the buret stopcock and make sure it is closed. Fill the reservoir bottle(s) with 0.0100 N
     H2SO4 and fill the buret by adjustment of the 3-way stopcock to the proper position. During
     analysis, cap the reservoirs with beakers. For overnight and longer storage, stopper the
     bottles with the appropriate sized stoppers. A regular 25- or 50-mL buret may also be used
     with a Nalgene self zero buret kit #364-2550.

 (3)  The filtration system should also be set up in advance. Clamp a 500-mL filter flask (serves
     as a trap) to a ring stand stabilized by a lead brick. Attach a piece of vacuum tubing to the
     vacuum source and to a 3-way connector. To one branch of the connector attach a short
     piece of vacuum tubing leaving the end open. Another piece of vacuum tubing must
     connect the remaining branch of the connector with the side arm of the 500-mL filter flask.
     Insert a 5-6 in. heavy-walled tube through the one-hole #7 stopper. Seat the stopper with
     tube in the flask and attach a piece of vacuum tubing to the tube. Connect an extension ring
     support to the stand (used  for holding the Buchner funnel when not in use). Insert the
     Buchner funnel through a  #6.5 neoprene stopper,  set it in the ring stand,  and cover the
     funnel with a 90-mm or larger watch glass.

 (4)  Insert the lower end of a condenser into a filter adapter or stopper that fits the Phillips
     beakers snugly. Connect the water supply to the bottom side arm of the condenser using a
     length of tubing. Also attach a drainage line (which runs to a sink) to the upper side arm of
     the condenser.
                                      CAUTION

           Securely fasten all connections using hose clamps, condenser
           hose clips or wire, to prevent the hoses from coming loose,
           causing flooding (especially dangerous around electrical
           equipment).
     Clamp the condenser to a ring stand, turn the water on, and adjust the flow rate.

 (5)  At the beginning of each day's analysis, fill the 2,000-mL Erlenmeyer flask with deionized
     water, add a stir bar, place a watch glass  over the mouth of the flask, and set the flask on a
     hotplate.
                                      WARNING

           A stir bar or glass bead must be placed in the flask to prevent
           violent bumping.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                   January 2016    Page 7 of 12

       Set the heat on high until the water begins to boil. After boiling starts, turn the heat down
       sufficiently to just maintain boiling. Also, at this time set a hot plate/stirrer by the ring
       stand with the condenser and adjust the heat setting to high. Set the third hotplate (to be
       used during the cleanup step) in a convenient place and set on high heat. Generally, a 45-
       to 60-min warmup period is required for the water and hotplates.

 A.4.4    pH Meter Calibration and Electrode Storage

 (1)  Calibrate the pH meter at the beginning and end of a run as described in the instrument
     manual and record the calibration data.

 (2)  Store electrodes in the 0.1 M KC1/0.0001 M HC1 solution described in Section 4.2, Step 3.
     For short-term storage and between samples the electrode may be placed in a beaker
     containing approximately 100 mL of storage solution. For long-term storage (weekend or
     longer) slip the electrode into an O-ring sealed electrode storage bottle approximately
     three-fourths full of storage solution and tighten the cap to seal the electrode body to the
     bottle.

 A.4.5   Analysis of Glass Fiber Filters
                                       WARNING

           Protective equipment must be worn to prevent serious burns. The
           sample is hot and caustic and will adhere to the skin.
 (1)  Clamp a clean, 250-mL filter flask to the filtration stand and connect the tubing from the
     vacuum trap to the flask side arm. Set the Buchner funnel on top of the filter flask, place a
     single 7-cm Whatman 41 or 541 filter circle in the funnel using tweezers, and cover the
     funnel.

 (2)  Set the first Phillips beaker, containing filter squares, near the hotplate/stirrer and add a 2-
     in. magnetic stir bar.

 (3)  Pour 100 mL of boiling water from the 2-L flask into a graduated cylinder. (Use heat-
     resistant gloves to handle the hot glassware.)

 (4)  Set aside the aluminum weighing dish cover, place the Phillips beaker on the hotplate, and
     turn on the stirrer.

 (5)  Add the 100 mL of water to the Phillips beaker, begin timing the extraction, quickly lower
     the condenser into the beaker and clamp loosely.

 (6)  At approximately 19 minutes and 30 seconds after the start of the extraction, don the face
     mask, turn on the vacuum, and put on heat-resistant gloves.
 (7)  At about 19 minutes and 55 seconds quickly lift the condenser out of the Phillips beaker
                                               2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters

     and clamp in a raised position.
January 2016    Page 8 of 12
                                        NOTE
           The stirrer may be stopped at this time.
 (8)  At exactly 20 minutes, using heat-resistant gloves, remove the Phillips beaker from the
     hotplate, swirl, and pour the contents rapidly, but carefully, through the Buchner funnel.
     Set the beaker on the bench and apply vacuum until the filter pad is just dry on top. The
     unconnected length of vacuum tubing may be used to regulate the vacuum strength by
     placing a finger over the hole in the tubing or by leaving it open.

 (9)  Remove the vacuum  tubing and the funnel from the filter flask. Pour the hot extract back
     into the Phillips beaker carefully washing down any residual pulp. Reclamp the filter flask,
     replace the funnel, and reattach the vacuum tubing. Swirl and pour the extract through the
     filter. Apply vacuum as needed.
                                        NOTE

           Steps 8 and 9 should be completed in 2 minutes or less. Typical
           times have been 1  minute and 30 seconds to 1 minute and 50
           seconds.
 (10)    Remove the fact shield, detach the vacuum tubing, set aside the Buchner funnel, place
        the filter flask on the titration stand, and remove gloves.
                                        NOTE

           Before titrating the extract just finished, a more efficient use of
           time may be made by starting the next sample at this point as
           follows: Remove the stir bar and wipe it clean of filter material
           adhering to it. Immediately start another extraction by performing
           Steps 2 through 5 on the next sample.
 (11)    Rinse the pH electrode with deionized water, shake off excess, pat dry with a laboratory
        tissue, and place in the filter flask. Push pH button (meter in Manual Temperature
        Compensation mode, set to 25°) and take an initial pH reading; this reading should be
        between pH 9 and 11 for glass filters and between 7 and 9 for quartz. Swirl the flask
        and begin titration.

 (12)    Titrate samples with 0.0100 N JfeSCU. Add the titrant in increments of several
        milliliters at a time until the pH drops below 8. At that point slow the addition rate to
        smaller increments of 0.1-0.5 mL of titrant. Stop the titrant flow when the pH reaches
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                  January 2016    Page 9 of 12

        7.00 and stop swirling; the pH should rise by 0.1-0.2 units. If the final calm pH is
        greater than 7.00, add enough titrant to compensate. If the titration cannot be stopped at
        pH 7.00 ±0.02, record the calm pH and the volume of titrant for points around pH 7 and
        calculate the extract volume of titrant required to reach pH 7 by linear regression or
        extrapolation methods.

 A.4.6   Analysis of Quartz Fiber Filters

 In order to reduce background alkalinity retained in the funnel, rinse the Buchner funnel once
 with boiling water before each analysis of a quartz filter. The alkalinity of quartz filters is more
 than an order of magnitude lower than that of glass filters; therefore, a lower background is
 required.

 (1)   Clamp in place a clean, dry, 250-mL filter flask and a 250-mL filter flask labeled "rinse."

 (2)   Seat the funnel on the rinse flask.

 (3)   Fill a graduated cylinder with 100 mL boiling deionized water, pour the water through the
      funnel and cover the funnel with a watch glass.

 (4)   through (9)  Same  as Steps 1-5 for glass fiber filters.

 (10) During the extraction, remove the cover from the funnel, pat the funnel bowl and tip dry
      with laboratory tissues, move the funnel to the clean dry flask, using forceps place a 7 cm
      Whatman 41 or 541 filter circle in the funnel and recover. Remove the rinse flask and
      drain.

 (11) through (17)  Same as Steps 6-12 for glass fiber filters.

 A.4.7   Analysis of Blanks

 Blank flasks are empty flasks that do not contain any filter squares.

 (1)   through (11) Same  as Steps 1-11 for glass fiber filters.

 (12a)  FOR BASIC BLANKS: If the initial pH is above 7, titrate slowly with the FfeSCM used
        for samples. See Step  12 for glass fiber filters.

 (12b)  FOR ACIDIC BLANKS: If the initial pH is below 7, titrate slowly with the 0.005 N
        NaOH from a 10-mL buret, instead of FhSO/t.  Slow the titrant flow when the pH goes
        above 6 and stop the flow when pH reaches 6.8-6.9; when swirling is ceased the pH
        should rise 0.1 to 0.2 units.  See Step 12 for glass fiber filters.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                  January 2016    Page 10 of 12

 A.4.8   Cleanup after Analysis

 (1)   Wipe and rinse the stir bars, then place in a clean dry place.

 (2a)  FOR GLASS FILTERS: Get the next clean filter flask, mount it on the ring stand, and
      attach the vacuum line. Remove the used filter with forceps, clean the funnel with
      laboratory tissues, seat it on the filter flask, add a fresh Whatman 41 or 541 filter with
      forceps, and cover the funnel.

 (2b)  FOR QUARTZ FILTERS: Mount the next filter flask and the rinse flask on the ring stand.
      Attach the vacuum line to the filter flask. Remove the used filter with forceps, clean the
      funnel with laboratory tissues, seat the funnel on the rise flask and cover.

 (3)   After each titration, rinse the pH electrode into a waste beaker, then with a laboratory
      tissue wipe off any condensation on the electrode, and place it in the  storage beaker.

 (4)   As time permits, rinse the Phillips beakers and filter flask thoroughly with a stream of
      deionized water. Then fill each twice with fresh deionized water discarding the water each
      time. Fill each a third time with fresh deionized water, add a small Teflon-coated stir bar
      or clean glass bead and place on a hotplate to boil (as time and space permit).
                                      WARNING

           Be sure to use heat-resistant gloves or tongs when handling the
           hot glassware to prevent burns.
      After boiling for at least 5 minutes, discard the water, taking care not to lose the stir bar,
      drain, and allow to dry thoroughly before reusing. Once per week, or more often as
      needed, scrub the sides and spout of the Phillips beaker with a clean brush, then rinse
      thoroughly and boil as above. Store the flask in a clean dry place.

 A.5.   Calculations

 (1)  If the titration was not stopped at exactly pH 7.00, then the titrant volume corresponding to
     pH 7.00 is calculated by extrapolation or linear regression.

 (2)  Calculate the microequivalents for the samples and the blanks using the following
     equation:

              microequivalents = (mL of titrant) x (normality of titrant) x 1,000.

      The sign of the blank (or sample) is negative if the titrant was NaOH and positive if the
      titrant was FhSO/t.
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2.12.Appendix A—Measuring Alkalinity of Filters                  January 2016     Page 11 of 12
 (3)  Next average the blanks for the day:
                                    blankl + blank2
                    average blank =	~	  (microequivalents)
 (4)  Then calculate alkalinity in microequivalents per gram of sample using the following
     equation:

                        jueq   jueq of sample - jueq of average blank
                         g           gram weight of sample
 (5)  If units of microequivalents per square centimeter are required:

                       /jeq   jueq      weight of entire filter (g)
                       cm2    g    surface area of entire filter (cm2)


 (6)   Sample calculation

      A filter weighing 4.0432 g was found to have an area of 517.2 cm2. A 1.005-g filter
      sample was extracted and the extract titrated with 0.0100 N H2SO4. The volume 10.99 mL
      of titrant corresponded to pH 7.08 and 11.12 mL of titrant corresponded to pH 6.92. Both
      blanks  required H2SO4 and had values of 1.0 and 0.8 microequivalents, respectively.
      Alkalinity is calculated as follows:

      Step 1 using linear regression:
         The estimated volume of titrant at pH 7.00 in this case equals 11.06 mL; or

      Step 1 using extrapolation
         The estimated volume of titrant also equals 11.06 mL.

      Step 2
         ueq = (11.06 mL) x (0.0100 N)  x  1,000
             = 110.6 microequivalents
      Step 3
                             .       ,,  ,   1.08+0.8
                            Average blank =	  =0.9 peg
                                              2.12-PM25 Reference/Class I Equivalent Methods

-------
2. 12. Appendix A — Measuring Alkalinity of Filters                 January 2016    Page 12 of 12


Step 4

                                            (110.6- 0.9) neq
                         alkalinity fiueq/g) =     j_005 g

StepS
                                        4.0432 g   109.2 ueq         ueq
                  alkalini ty  -   = c>?  ..  * 2 x - ^ = 0.854 ^
                              J cm2   5 17. /2cm2       g             cm2
 A.6   Quality Control

 Prepare a total of nine Phillips beakers for analysis each day; leave two empty as blanks. The
 sample analysis order is: two samples, one blank, three samples, one blank, two samples. Two
 samples should be from filters previously analyzed. Number the Phillips beakers and filter
 flasks from 1 to 9 and rotate the position each day so all glassware pairs eventually are used for
 a blank analysis. An example of the glassware and sample order for a week is given in Table A-
 3.
                  TABLE A-3. GLASSWARE AND SAMPLE ORDER

Day 1
Glass Sample
1
2
3
4
5
6
7
8
9
S-1
S-2
B-1
S-3
S-4
S-5
B-2
S-6
S-7

Day 2
Glass Sample
2
3
4
5
6
7
8
9
1
S-8
S-9
B-3
S-10
S-1
S-11
B-4
S-1 2
S-5

Glas
s
3
4
5
6
7
8
9
1
2
Day 3
Sample
S-1 3
S-1 4
B-5
S-1 5
S-9
S-1 6
B-6
S-1 7
S-1

Day 4
Glass Sample
4
5
6
7
8
9
1
2
3
S-1 8
S-1 9
B-7
S-20
S-1 6
S-21
B-8
S-22
S-1 3

Day 5
Glass Sample
5
6
7
8
9
1
2
3
4
S-23
S-24
B-9
S-25
S-1 8
S-26
B-10
S-27
S-4
                                             2.12-PM25 Reference/Class I Equivalent Methods

-------
Page Intentionally Left Blank

-------
O  ^H

O  O



£  ^


ffl   S3









PH



 O
 o
fin
"S
&   o  O
"^  '55  Z;
 c  -c1

        PL,
    X"  ao

     c   §
     o
    *-^
     o
     (L)
    -4—»

     s
    PH
 t«  jj
 (L)   C

"oS   ^


^   I   >,
' (^J   Q   f^

^  .b   c
.ti   >   (u
 e   c   oo

£  W  
-------