QUALITY ASSURANCE PROJECT PLAN (QAPP)

FORTHE

REMOVAL INVESTIGATION

AT THE

WHITE SWAN LAUNDRY AND CLEANERS SITE

WALL TOWNSHIP, MONMOUTH COUNTY, NEW JERSEY

Project Officer's-Signature: ^ C x ^ ^ - ^ g ^ ^ o . Date: 1

Project Officer's Name: Diane Salkie, EnvipjHsental Scientist

Project Quality Assurance Officer's Signature: ~ j ^ ~ ^ ^ J U ^ ^ ^ l M e : ^

Project Quality Assurance Officer's Name: Pat Sheridan, QA Officer

Date Prepared: December 14, 2001

TABLE OF CONTENTS

QAPP Element Page

1.0 PROJECT DESCRIPTION , 1 1.1 Project Definition/Background 1 1.2 Project/Task Description ...2

2.0 PROJECT ORGANIZATION AND RESPONSIBILITY. 3 2.1 Project/Task Organization 3 2.2 Documentation and Records 4

3.0 QA OBJECTIVES FOR MEASUREMENT DATA (PARCC) 4 3.1 Quality Objectives and Criteria for Measurement Data 4

3.1.1 Analytical and sample collection precision 5 3.1.2 Analytical and sample collection accuracy 5 3.1.3 Data representativeness 5 3.1.4 Data completeness 6 3.1.5 Data comparability 6

4.0 SAMPLING PROCEDURES 7 4.1 Sampling Process Design ; 7 4.2 Sampling Methods Requirements 8

4.2.1 Standard operating procedures 8 4.2.2 Sample collection methodology 8 4.2.3 Sample containers, volume, preservation, and holding times....! 8 4.2.4 Field measurement data collection 9 4.2.5 Sampling equipment decontamination 9 4.2.6 Management of investigative-derived wastes (IDW) 9

5.0 SAMPLE CUSTODY '. 9 5.1 Special Training Requirements or Certifications 9 5.2 Sample Handling and Custody Requirements : 10 '

5.2.1 Sample handling and shipment 10 5.2.2 Sample custody procedures 11

6.0 CALIBRATION PROCEDURES AND FREQUENCY....: 12 6.1 Instrument Calibration and Frequency :. 12

7.0 ANALYTICAL PROCEDURES 12 7.1 Analytical Methods Requirements 12

8.0 DATA REDUCTION, VALIDATION, AND REPORTING 12 8.1 Data Review, Validation, and Verification Requirements 12 8.2 Validation and Verification Methods ; 12" 8.3 Data Acquisition Requirements : 12 8.4 Data Quality Management 12

9.0 INTERNAL QUALITY CONTROL CHECKS AND FREQUENCY 14 9.1 Quality Control Requirements 14

9.1.1 Data precision 14 9.1.1.1 analytical precision ; 14 9.1.1.2 sample collection precision 14

9.1.2 Data accuracy '. 15 9.1.2.1 analytical accuracy.. 15 9.1.2.2 sample collection accuracy.: 15

9.1.3 Data representativeness 15 9.1.4 Data comparability 16 9.1.5 Data completeness 16

TABLE OF CONTENTS (Continued)

QAPP Element Page

10.0 PERFORMANCE AND SYSTEMS 16 10.1 Assessments and Response Actions '. 16

11.0 PREVENTIVE MAINTENANCE 16 11.1 • Instrument/Equipment Testing, Procedures and

Scheduled Inspection, and Maintenance Requirements 16 11.2 Inspection/Acceptance Requirements for Supplies and Consumables 17

12.0 SPECIFIC ROUTINE PROCEDURES MEASUREMENT PARAMETERS INVOLVED 17 12.1 Reconciliation with Data Used to Assess PARCC for Quality Objectives Measurement 17

13.0 CORRECTIVE ACTION 17 13.1 Assessments and Response Actions 17

14.0 QA REPORTS TO MANAGEMENT 18 14.1 Distribution List 18 14.2 Reports to Management 18

ii

LIST OF APPENDICES

Appendix A - Site Maps

Appendix B - Eastern Research Group (ERG). March 2000. Support for NMOC/SNMOC, UATMP and PAMS Networks, Contract No. 68-D-99-007. Quality Assurance Project Plan. Morrisville, North Carolina

Appendix C - U.S. Environmental Protection Agency (EPA). January 1999. Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) from the Compendium.ofMethods for the Determination ofToxic Organic Compounds in Ambient Air. Second Edition. Center for Environmental Research Information. Office of Research and Development. Cincinnati, OH

Appendix D - U.S. EPA. July 1995. Environmental Response Team (ERT) Standard Operating Procedure (SOP) #1704: SUMMA Canister Sampling.

Appendix E - U.S. EPA. September 1994. Environmental Response Team (ERT) Standard Operating Procedure (SOP) #170: Sample Documentation.

Appendix F- ERG's Method Detection List

Appendix G - Example Questionnaire, Example Canister Field Data Sheet and Example Chain of

Appendix H - Resident Instructions

I

ii

1.0 Project Description

1.1 Problem Definition/Background

The White Swan Laundry and Cleaners site (site) encompasses three different potential responsible parties (PRPs) in three areas. They are the former White Swan Laundry and Cleaners on 1322 Sea Girt Avenue in Wall Township; the Gulf Service Station on 1324 Sea Girt Avenue in Wall Township; arid the former Sun Cleaners on 2213 Route 35 in Wall Township. However, the name of the site is White Swan Laundry and Cleaners. See Map 1 of Appendix A for a map of the site location and Map 2C for locations of the three areas of concern.

The White Swan Laundry and Cleaners, located in Block 706, lot2, began as an ice cream parlor, the Big Scoop, in 1957. The Big Scoop was sold to Mr. Harry B. (See Map 2 of Appendix A for a site location) White in 1964 who began operations of White Swan Laundry and Cleaners and continued for 18 years. Mr. White sold the business and property to Charles J. and Mary M. Mahoney in 1982. They held the business for less than one year when they sold the property to Ocean County National Bank in 1983. The bank demolished the original building and constructed a new one. The property has continued as a bank ever since and was connected to the public sewer system in 1986 when Summit Bank purchased the property. Prior to 1986, the site used a septic system for all of its discharges. The site is in a commercial/residential area. A Gulf Service Station is located to the west, a motel is located to the north and a convenience store/hair salon is to the east. Across the street to the south is a bank and a strip mall. Residential properties are located to the northeast/east. New Jersey Department of Environmental Protection (NJDEP) conducted an investigation of the property in January of 2000 and confirmed a release of PCE to the soil and groundwater. See Map 3 of Appendix for a diagram of the site.

The Gulf Service Station, located in Block 706,- lot 1, was purchased by the Gulf Oil Corporation in 1951 as an undeveloped property from Lawrence and Helen Edwards. See Map 2A of Appendix A for a site location. From 1951 until 1986, The Gulf Oil Corp. maintained ownership of the property, but leased it to several operators who distributed gas and conducted auto repairs. In 1986, the Gulf Oil Corp. was purchased by Chevron USA, Inc. and in 1986 the property was purchased and is still owned by Cumberland Farms, Inc., but operates under the Gulf Service Station name. Prior to 1986, when the property was connected to public sewer, the facility used a septic system, located in the northeast corner of the property. Currently the facility utilized four underground storage tanks for the distribution of gasoline. There is an underground waste oil tank on the northwest side and an underground fuel oil tank in the back of the building. An underground kerosene tank was removed in 1998. NJDEP personnel discovered an unregistered underground waste tank on the east side of the building. The original gasoline tanks were removed in 1984 and replaced with fiberglass tanks. NJDEP conducted an investigation of the property in 2000 and confirmed PCE in the on-site soil and groundwater. See Map 3 A of Appendix A for a diagram of the site.

1

The first records of Sun Cleaners, located in block 807, lot 1, show that it began as Circle Dry Cleaning Corporation in 1960. See Map 2B of Appendix A for a site location. They operated a dry cleaning operation for 22 years. In 1982, Sun Cleaners began operating at the facility and lasted for 9 years. In 1991, the owner, Sylvia Harte, ceased all on site dry cleaning operations and only operated as a drop-off and pick-up of dry cleaning materials. The facility is a two-story building with commercial space on the first floor. The southern portion of the first floor was used as a dry cleaning drop off service and the northern section was used for the dry cleaning of clothes. Although the facility ceased operations, the equipment remains in the building. A poly tank is located on the roof above the dry cleaning section of the building. The tank reportedly held water used in dry cleaning operations. The second floor of the building consists of two apartment units which have not been occupied for years due to unsafe conditions. A septic system was in operation until the early 1980s when the facility was connected to a public sewer system. An underground storage tank is located on the south side of the property. Located at the north end of the building are: a 55-gallon drum, 30-gallon drum and a separator discharge pipe. See Map 4 of Appendix A for a diagram of the site.

NJDEP conducted several investigations in 2000 throughout the site, specifically ground water sampling. Due to high levels of PCE and TCE in the groundwater, NJDEP also sampled air in the basements of residents in the area. This investigation proved some of the homes to have PCE, TCE and benzene contamination in the vapors of the basements. The Division of Environmental Science and Assessment (DESA), Hazardous Waste Support Branch (HWSB), Superfund Contract Support Team (SCST) has been requested by the Environmental Remedial and Response Division (ERRD) to continue sampling the vapors in the basements of the residential houses in the surrounding vicinity of the site.

Project/Task Description:

The purpose of this removal assessment is to collect valid data which are necessary and efficient to verify that contaminants exist in the residential basements surrounding the site. The sampling event will also determine whether or not an immediate threat to human health or the environment exists. The scope of the removal assessment is to:

• assess the extent of contamination in the basements of residents; and

• delineate the specific organic contaminants in the vapors in the basements of residents.

All analysis of the air samples collected during this sampling event will be performed by the Eastern Research Group, Inc. (ERG). This sampling event will use the ERG contract to supply the SUMMA™ canisters and submit them to their own laboratory for analysis. A copy of the quality assurance project plan for the contract entitled Support For NMOC/ SNMOC, UATMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan can be found as Appendix D. This document will be referred to as the contractor's or ERG QAPP throughout this document. According to the QAPP, the relevant program for this sampling event is the Urban Air Toxic Monitoring Program (UATMP)

The purpose and scope of this QAPP is to specify the details related to the collection, analysis and validation of the samples collected by the USEPA Region 2, DESA, HWSB, SCST from December 20, 2001 until January 20, 2002. The activity schedule is as follows:

ACTIVITY DATE

Date of the request which initiates the project. December 06, 2001

Review and Background information December 11, 2001

Date by which the project plan will be submitted to all interested parties.

December 14, 2001

Obtain site access Prearranged by ERRD

Date by which comments on the plan are to be received by the project officer.

December 19, 2001

Date(s) of the field reconnaissance, December 20, 2001

Date(s) of the field sampling activities. December 20, 2001 -January 20, 2002

Date(s) the samples will be submitted to the laboratory for analysis.

All samples will be shipped within 24 hours of collection.

Date(s) by which all analyses are to be completed and the data submitted to the project officer.

30 days.

Date(s) the data will be entered into STORET or other computerized systems.

Not applicable.

Date of the completion of the draft interim/final project report. (Sampling Trip Report)

Within one week of the end of the sampling event

Date by which the reviewer's comments on the report(s) must be received.

Not applicable.

Date for completion of the peer review process. Not applicable.

Date for the issuance of the final project report. Within two weeks of receipt of validated analytical data.

The primary use of the data collected will be to determine the extent of air contamination, evaluate potential health risks, and determine environmental impacts. The samples results will be submitted to Agency for Toxic and Disease Registry (ATSDR) who will determine whether the contamination is significant enough to cause an adverse effect on human health.

2.0 PROJECT ORGANIZATION AND RESPONSIBILITY

2.1 Project/Task Organization

The following is a list of key personnel and their corresponding responsibilities. Due to the work breakdown structure of the project, an organization list is provided instead of a concise organization chart.

PROJECT PERSONNEL RESPONSIBILITY

Diane Salkie, Project Officer DESA/HWSB Superfund Contract Support Team

Project Management/ Sampling Operations/ Field Support

Keith Glenn, Environmental Scientist DESA/HWSB Superfund Contract Support Team

Sampling Operations/ Field Support

Michael Mercado, Environmental Scientist DESA/HWSB Superfund Contract Support Team

Sampling Operations/ Field Support

Pat Sheridan, Project Quality Assurance Officer DESA/HWSB/HWSS

Report QA

ERG provided Laboratory Laboratory Analysis, Laboratory QC, Data Processing Activities, Data Quality Review

Not Applicable Performance Auditing

Not Applicable Systems Auditing

DESA/Hazardous Waste Support Branch Overall QA

Thomas Budroe, On-Scene Coordinator ERRD/RAB

Overall Project Coordination

2.2 Documentation and Records

The data collected for the sampling activities will be organized, analyzed, and summarized in a final project report that will be submitted to the OSC according to the Project Schedule. The report will be prepared by the project officer and include appropriate data quality assessment. Standard methods and references will be used as guidelines for data reduction and reporting. All SOP data generated by the laboratory will be reported in standard deliverable format.

3.0 QUALITY ASSURANCE (QA) OBJECTIVES FOR MEASUREMENT DATA (PARCC)

3.1 Quality Objectives and Criteria for Measurement Data

To assess data quality, PARCC (Precision, Accuracy, Representativeness, Completeness, and Comparability) parameters will be utilized. This is an integral part of the overall monitoring network design. Precision and accuracy are expressed in purely quantitative terms. The other parameters are only expressed using a mixture of quantitative and qualitative terms. All of these parameters are interrelated in terms of overall data quality and they may be difficult to evaluate separately due to these interrelationships. The relative significance of each of the parameters depends on the type and intended use of the data being collected. Therefore, these essential data quality elements are delineated as follows.

4

3.1.1 Analytical and sample collection precision

The measure of replicate precision is the absolute value of the difference between replicate measurements of the sample divided by the average value and expressed as a percentage as follows:

Percent difference = \X{ - X,| x 100

x ; where: X, - First measurement value

X 2 - Second Measurement value X - Average of the two values

Factors that affected the precision of the measurement are: molecular weight, water solubility, polarizability, etc. A primary influence is the concentration level of the compound. A replicate precision value of 25 percent can be achieved for each of the target compounds. For more information, refer to Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) which can be found as Appendix C.

3.1.2 Analytical and sample collection accuracy

A measurement of analytical accuracy is the degree of agreement with audit standards. It is defined as the difference between the nominal concentration of the audit compound and the measured value divided by the audit value and expressed as a percentage as follows:

Audit Accuracy, % = Spiked Value - Observed Value X 100 Spiked Value

For more information, refer to Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) which can be found as Appendix C. As per Method TO-15, the performance criteria for audit accuracy should be within 30 percent for concentrations normally expected within contaminated ambient air.

3.1.3 Data representativeness

As previously discussed, data representativeness will be assessed by collecting field . replicate samples. The field replicates are by definition equally representative of a given point and space and time. Representativeness is a qualitative parameter which is dependent upon the proper design of the sampling program and proper laboratory protocol. Therefore, data representativeness will be satisfied by ensuring that:

5

The sampling program is followed according to:

U.S. EPA (Environmental Protection Agency). October 1989. Region I I CERCLA Quality Assurance Manual. Final Copy, Revision 1. Division of Environmental Services and Assessment, Edison, NJ.; and

U.S. EPA. December 1995. Superfund Program Representative Sampling Guidance. OSWER Directive 9360.4-10: Interim Final. EPA/540/R-95/141. Office of Emergency and Remedial Response (OERR). Washington, D.C.

Proper sampling techniques are used in accordance with:

U.S. EPA. Environmental Response Team (ERT) Standard Operating Procedure (SOP) #1704: Summa Canister Sampling; revised July 1995. The SOP is enclosed in Appendix D.

Proper analytical procedures are followed and holding times of the samples are not exceeded in the laboratory according to:

U.S. EPA. January 1999. Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) from the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air. Second Edition. Center for Environmental Research Information. Office of Research and Development. Cincinnati, OH which can be found as Appendix C.

3.1.4 Data completeness

Data completeness will be expressed as the percentage of valid data obtained from measurement system. For data to be considered valid, it must meet all the acceptable criteria including accuracy and precision, as well as any other criteria specified by the analytical method used. Therefore, all data points critical to the sampling program in terms of completeness will be 100% validated by ERG according to Section 15 of ERG's QAPP which can be found as Appendix B. With 100% validation, the rationale for considering data points non-critical is not required.

3.1.5 Data comparability

To ensure data comparability, sampling and analysis for all samples will be performed using standardized analytical methods and adherence to the quality control procedures outlined in the methods and this QAPP. Therefore, the data will be comparable.

6

4.0 SAMPLING PROCEDURES

4.1 Sampling Process Design

As part of the Removal Assessment process, U.S. EPA Region IIDESA/HWSB/SCST personnel will collect air samples in the basements of houses in the area of the White Swan Laundry and Cleaners site. Samples will be collected with SUMMA™ canisters. SUMMA™ Canister sampling will follow methods as described in U.S. EPA/ERT SOP #1704, Appendix D.

The sampling design for the site, including the rationale for sample frequency, location, and depth was predetermined by the On-Scene Coordinator (OSC). For the purposes of this sampling event, sample location selection was determined by selecting locations of in the area suspected ground water contamination. Each location to be sampled will be chosen by the OSC prior to sampling. The resident will be notified prior to the sampling day by a telephone call and an instruction page through the mail. A copy of this page can be found as Appendix H. A map of the area can be found in Appendix A. A detailed description of sample collection methodology is presented in Section 4, Sub-section 2, Part 2: Sample Collection Methodology.

A total of one hundred and seventy three (173) samples will be collected from one hundred and fifty (150) homes and eight (8) background locations. The one hundred and seventy three (173) samples also include a maximum of 10% QA/QC (field replicate samples). All samples will be analyzed for Title I I I Clean Air Amendment List - volatile organic compounds (VOC) which can be found in Appendix F. All samples will be collected by U.S. EPA personnel and then sent to ERG contractors. The contractors are providing the canisters and submitting the samples to their own laboratory who will analyze the samples according to TO-15 which can be found as Appendix B.

The sampling and analysis protocol is listed as Table 1 on page 8.

7

TABLE 1 \ \ hite Swan Laundry and Cleaners Site

Removal Investigation Sampling and Analysis Protocols

Sample Type

Number of Samples

Matrix Parameter/Fraction Sample Container

; Sample Preservation

Analytical Method 1

Method Detection

Limit

Holding Time

Residential Basements

165 Air Title III Clean Air Amendment List

Volatile Organic Compounds (VOCs)

(1) SUMMA Canister

TO-15 0.04 - 0.26 ppbv

30 days

Background Samples

8 Air Title III Clean Air Amendment List

Volatile Organic Compounds (VOCs)

(1) SUMMA Canister

TO-15 0.04 - 0.26 ppbv

30 days

Legend - ' r - 1 j < ' U S : EPA. January .1999. Compendium Method TO-15: Determination.ofVolatile•OrganicCompounds(VOCs)>:.m Air Collected in Specialty-""-Prepared Canisters and Analyzed.byGas Chromatography/Mass Spectrometry(GC/MS) from the Compendium of Methods for the Determination ofcToxic OrganicCompounds in Ambient'Air. Second Fdition

4.2 Sampling Methods Requirements

4.2.1 Standard operating procedures

As previously stated, all sampling will be in accordance with the U.S. EPA Region II CERCLA Quality Assurance Manual; and U.S. EPA Superfund Program Representative Sampling Guidance OSWER Directive 9360.4-10, Interim Final, EPA/540/R-95/141, Office of Emergency and Remedial Response (OERR), Washington, D.C. Furthermore, the specific Standard Operating Procedure (SOP) utilized for air sampling, as presented in Appendix D, is the U.S. EPA ERT SOP #1704: Summa Canister Sampling.

4.2.2 Sample collection methodology

All samples including QA/QC samples will be collected by personnel from the US EPA Region IIDESA/HWSB/SCST from the residential basements in the area of the White Swan Laundry and Cleaners site. The total number of samples includes: one hundred and fifty (150) samples in addition to: up to fifteen (15) laboratory quality control samples (i.e. field duplicates) and eight (8) background samples. Samples will be collected by placing a SUMMA™ canister in the residential basement, setting the valve for the appropriate amount of time and retrieving the canister after 24 hours.

4.2.3 Sample Containers, Volume, Preservation, and Holding Times

Sample container type, volume, preservation, and holding times are dependent upon analytical parameter and fraction and are matrix specific. The following table outlines the sample container type, volume, preservation, and holding times for samples to be collected

. on-site.

8

Analytical Parameter/Fraction i ,. Sample Container . ..>:

,, Required S Sample {• 'Volume

Sample Preservation , HoldingiTime i

Clean Air Amendment List - VOC

(1) SUMMA™ canister 6 Its. 30 days to analyze

4.2.4 Field measurement data collection

A photo-ionization detector (PID) will be used for the health and safety of the samplers. Canister Sample Data Sheets, Questionnaires and the field notebook will be completed for each sample collected. The Questionnaire will record sample location; residential information; time of sample drop off and pick up; conditions in the room; laboratory sample number; laboratory sample analysis and sample collection notes and/or observations. An example of the Questionnaire is presented in Appendix G. The Canister Sample Data Sheet will be provided by ERG and records the sample location, sampling period, initial and final pressure and comments. An example of this data sheet can also be found in Appendix G. The field notebook will be completed as provided for in Section 8.4: Data Quality Management of the QAPP.

4.2.5 Sampling Equipment Decontamination

Air samples will be collected using summa canisters. ERG will perform decontamination of the canister prior to sending them to U.S. EPA. The SUMMA™ canisters will be cleaned according to:

• Support For NMOC/SNMOC, UATMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan (Appendix B);

• U.S. EPA ERT SOP #1703: Sample Documentation (Appendix E); and

• U.S. EPA Region IICERCLA Quality Assurance Manual.

4.2.6 Management of Investigative-Derived Wastes (IDW)

The wastes that are anticipated on being generated during this sampling event are personnel protective equipment (i.e. goggles, booties, etc.). The personnel protective equipment will be double-bagged and properly disposed of in on-site solid waste roll-off or off-site in properly designated containers.

5.0 SAMPLE CUSTODY

5.1 Special Training Requirements/Certification

To perform the operations of this sampling event, SCST will be dealing with the removal activities on-site. This can imminently expose SCST personnel to potential occupational environmental hazards. As a result, it is important for SCST field personnel to be familiar with:

9

• Identifying methods and procedures for recognizing, evaluating and controlling hazardous substances.

• Identifying concepts, principles, and guidelines to properly protect SCST field personnel.

• Discussing regulations and action levels to ensure the health and safety of SCST field oversight personnel.

• Discussing the fundamentals needed to develop organizational structures and standard operating procedures to mitigate potential environmental hazards.

• Demonstrating the selection and use of dermal and respiratory protective equipment.

• Demonstrating the selection and use of direct-reading air monitoring instrumentation (if applicable).

In practice, not all of the potential environmental hazards which may be inherent to a site can be readily anticipated. To mitigate these circumstances, SCST field personnel must learn, follow, and enforce the published rules governing occupational health and safety. In addition, they must maintain awareness and exercise common sense and good judgement when confronting possible unsafe situations. Consequently, all divisions and offices at the Edison facility are required to provide their staff with the necessary safety training and equipment to perform their assigned duties.

For SCST personnel, all training and certification requirements are to be undertaken in accordance with the protocols set forth in the 1995 "Edison Health and Safety Manual." Specifically, this requires completion of the forty (40) hour "Hazardous Materials Incident Response Operations" training pursuant to Occupational Safety and Health Administration (OSHA) regulation 29 CFR 1910.120 and U.S. EPA Order 1440.2. This is to be supplemented by completing the twenty four (24) hour OSHA sanctioned supervised on-site operations certification training. In conjunction, SCST personnel are also to maintain certifications for:

• The supplemental eight (8) hour annual health and safety refresher training. • Fit testing for atmosphere supplying respirators (Level B) and air purifying

respirators (Level C). • Enrollment in a physician authorized medical monitoring program.

5.2 Sample Handling and Custody Requirements

5.2.1 Sample handling and shipment

Canister Sample Data Sheets, a Questionnaire and the field notebook will be completed for each sample collected. All field and sample documents will be legibly written in indelible ink. Any corrections or revisions will be made by lining through the original entry and initialing the change. The Questionnaire will record sample location; residential information; time of sample drop off and pick up; conditions in the room; laboratory sample number; laboratory sample analysis and sample collection notes and/or observations. For reference, an example of the Questionnaire is presented as Appendix G. The Canister Sample Data Sheet will be provided by ERG and records the sample location, sampling period, initial and final pressure and comments. An example of this

10

data sheet can also be found in Appendix G. The field notebook will be used by field personnel to record all aspects of sample collection and handling, visual observations, and field measurements. The field notebook is a descriptive notebook detailing site activities and observations so that an accurate, factual account of field procedures may be reconstructed. The sample team or individuals performing a particular sampling activity are required to maintain a field notebook. This field notebook will be a bound weatherproof logbook that shall be filled out at the location of sample collection immediately after sampling. All entries will be signed by the individuals making them. At a minimum, the notebook will contain sample particulars including sample number, collection time, location, descriptions, methods used, daily weather conditions, field measurements, name of sampler(s), sample preservation, names of contractor/ subcontractor personnel, and other site-specific observations including any deviations from protocol.

The Canister Tag, found in Appendix G> also provided by ERG, will be securely affixed to each SUMMA™ canister and include only the sample identification number as per protocol. The sample tags will be secured to the canister itself. Custody seals will then be affixed around a bag surrounding each individual canister. Once sealed, samples will be placed back into the cardboard boxes that they were received in. Custody seals and strapping tape will then be affixed to the boxes.

Samples will be packaged and shipped in accordance with USEPA, Department of Transportation (DOT), and International Air Transport Association (IATA) procedures. All samples will be shipped within 24 hours of collection to the ERG office in North Carolina.

5.2.2 Sample custody procedures

Standard U.S.EPA Chain-of-Custody Procedures will be followed for all samples and be in accordance with the U.S.EPA Region I I CERCLA Quality Assurance Manual. The Chain of Custody Records will be maintained from the time of sample collection until final deposition. Every transfer of custody will be noted and signed for and a copy of the record will be kept for each individual who has signed it. The chain-of-custody records will include, at a minimum, sample identification number, number of samples collected, sample collection date and time, sample type, sample matrix, sample container type, sample analysis requested, sample preservation, and the name(s) and signature(s) of samplers and all individuals who have had custody. Sample labels will only include the sample identification number as per protocol to prevent any conflict of interest issues. Custody seals will demonstrate that a sample container or cooler has not been opened or tampered with. The sampler will sign and date the custody seal and affix it to the container and/or cooler in such a manner that it cannot be opened without breaking the seal.

11

6.0 CALIBRATION PROCEDURES AND FREQUENCY

6.1 Instrument Calibration and Frequency

Laboratory analytical equipment calibration will follow procedures as specified under U.S. EPA, Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) from the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, which can be found as Appendix C.

7.0 ANALYTICAL PROCEDURES

7.1 Analytical Methods Requirements

The analytical method, equipment and method performance requirements for analysis will be according to ERG's contract with the subcontracted laboratory. Refer to U.S. EPA, Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/ Mass Spectrometry (GC/MS) from the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, which can be found as Appendix C.

8.0 DATA REDUCTION, VALIDATION, AND REPORTING

8.1 Data Review, Validation and Verification Requirements:

Standard methods and references will be used as guidelines for data reduction and reporting. All data generated by the laboratory will be reported in standard deliverable format. ERG will be using a gas chromatograph (GC)/flame ionization detector (FID)/ mass selective detector (MSD) to analyze the samples for VOCs as stated in their QAPP which is based on TO-15. Due to ERG's vast experience with analyzing SUMMA™ canisters for VOCs, they have found this method to be the most precise and they are able to detect the compounds at unusually low concentrations. All data validation reports will be summarized according ERG's QAPP: Support For NMOO SNMOC, UATMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan, Section 15 which can be found as Appendix B.

8.2 Validation and Verification Methods

All data will be validated by ERG's generic Support For NMOO SNMOC, UATMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan, Section 15, which can be found as Appendix B.

8.3 Data Acquisition Requirements

Data acquisition from non-direct measurements such as data from databases or literature is not anticipated at this time. Therefore, this is not applicable.

12

Data Quality Management

All project data and information must be documented in a format that is usable by project personnel. This section of the QAPP describes how project data and information will be documented, tracked, and managed from their generation in the field to final use and storage in a manner that ensures data integrity and defensibility. All field and sample documents will be legibly written in indelible ink. Any correction or revisions will be made by lining through the original entry and initialing the change.

The following field and sample documentation will be maintained.

• The field notebook is a descriptive notebook detailing site activities and observations so that an accurate, factual account of field procedures may be reconstructed. The sample team or individuals performing a particular sampling activity are required to • maintain a field notebook. This field notebook will be a bound weatherproof logbook that shall be filled out at the location of sample collection immediately after sampling. All entries will be signed by the individuals making them. At a minimum, the notebook will contain sample particulars including sample number, collection time, location, descriptions, methods used, daily weather conditions, field measurements, name of sampler(s), sample preservation, and other site-specific observations including any deviations from protocol.

• Field data sheets, i.e., Questionnaire, Canister Field Data Sheet, and corresponding sample labels are used to identify samples and document field sampling conditions and activities. The field data sheets will be completed at the time of sample collection and will include the following: sample location; residential information; drop off and pick up time; sample environment description; laboratory sample number; laboratory sample analysis; and sample collection notes and/or observations. An example of the Questionnaire and the Canister Field Data Sheet are presented in Appendix G. Sample labels will be securely affixed to the sample container and include only the sample identification number as per Protocol.

• Sample tags will be securely affixed to the sample container and include only the sample identification number as per protocol to prevent any conflict of interest issues. The sample labels will be sealed to a bag surrounding the canister sample label integrity.

• The Chain of Custody Records will be maintained from the time of sample collection until final deposition. Every transfer of custody will be noted and signed for and a copy of the record will be kept for each individual who has signed it. The chain-of-custody records will include, at a minimum, sample identification number, number of samples collected, sample collection date and time, sample type, sample matrix, sample container type, sample analysis requested, sample preservation, and the name(s) and signature(s) of samplers and all individuals who have had custody-. An example of the chain of custody that will be used at this site can be found in Appendix G.

13

Custody seals will demonstrate that a sample canister or box has not been opened or tampered with. The sampler will sign and date the custody seal and affix it to the bag or box in such a manner that it cannot be opened without breaking the seal.

• Procedures are provided for project personnel to make changes, take corrective actions and document the process through Corrective Action Request Forms. Corrective action can occur during field activities, laboratory analysis, data validation, and data assessment. For further information, refer to Section 13.0: Corrective Action.

9.0 INTERNAL QUALITY CONTROL CHECKS AND FREQUENCY

9.1 Quality Control Requirements

As previously stated, to assess data quality, PARCC (Precision, Accuracy, Representativeness, Completeness, and Comparability) parameters will be utilized. These essential data quality elements are delineated as follows.

9.1.1 Data precision

Precision is defined as a measure of the reproducibility of individual measurements of the same property under a given set of conditions. The overall precision of measurement data is a mixture of sampling and analytical factors.

9.1.1.1 An aly tical precision

The measure of replicate precision is the absolute value of the difference between replicate measurements of the sample divided by the average value and expressed as a percentage as follows:

Percent difference = ]X, - X 2 | x 100 X

where: X! - First measurement value X 2 - Second Measurement value X - Average of the two values

Factors that affected the precision of the measurement are: molecular weight, water solubility, polarizability, etc. A primary influence is the concentration level of the compound. A replicate precision value of 25 percent can be achieved for each of the target compounds. For more information, refer to Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) which can be found as Appendix C.

Table 2 located on page 17 of this QAPP depicts the analytical precision for the analytical methods chosen in terms of estimated relative percent difference (RPD).

14

9.1.1.2 Sample collection precision

Sample collection precision will be assessed by collecting field replicate samples. The field replicates will be used to evaluate errors associated with sample heterogeneity, sampling methodology and analytical procedures. The analytical results from these samples will provide data on the overall measurement precision.

9.1.2 Data accuracy

Accuracy is defined as the degree of difference between measured or calculated values and the true value. The closer the numerical value of the measurement comes to the true value, or actual concentration, the more accurate the measurement is. It is difficult to measure accuracy for the entire data collection activity. Sources of error are the sampling process, field contamination, preservation, handling, sample matrix, sample preparation and analysis techniques.

9.1.2.1 Analytical accuracy

A measurement of analytical accuracy is the degree of agreement with audit standards. It is defined as the difference between the nominal concentration of the audit compound and the measured value divided by the audit value and expressed as a percentage as follows:

Audit Accuracy, % = Spiked Value - Observed Value X 100 Spiked Value

For more information, refer to Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) which can be found as Appendix C.

Table 2 on page 17 located in this QAPP depicts both the analytical precision and accuracy for the analytical methods chosen in terms of estimated percent recovery.

9.1.2.2 Sample collection accuracy

Method blanks will be used to monitor possible laboratory contamination.

9.1.3 Data Representativeness

Representativeness expresses the degree to which sample data accurately and precisely represent a characteristic of a population, parameter variations at a sampling point, or and environmental condition. Representativeness is a qualitative parameter which is most concerned with the proper design of the sampling program and proper laboratory protocol. The representativeness criterion is best satisfied by making certain that sampling locations are selected properly and a sufficient number of samples are collected. Therefore, data representativeness will be assessed by collecting field replicate samples. The field replicates are by definition equally representative of a given point in space and time.

15

In addition, as previously stated, data representativeness will be satisfied by ensuring that the sampling program is followed according to the U.S. EPA Region IICERCLA Quality Assurance Manual; and the U.S. EPA Superfund Program Representative Sampling Guidance for soil, Volume 1. Also, proper sampling techniques will be used in accordance with the U.S. EPA. Environmental Response Team (ERT) Standard Operating Procedure (SOP) #1704: Summa Canister Sampling. The SOP is enclosed in Appendix D.

9.1.4 Data Comparability .

Comparability is defined as the confidence with which one data set can be compared to another. Field and laboratory procedures greatly affect comparability. Therefore, to optimize comparability, sampling and analysis for all samples will be performed using standardized analytical methods and adherence to the quality control procedures outlined in the methods and this QAPP. Therefore, the data will be compared.

9.1.5 Data Completeness

Completeness is defined as the measure of the amount of valid data obtained from a measurement system compared to the amount that was expected to be obtained under normal conditions. Data completeness will be expressed as the percentage of valid data obtained from measurement system. For data to be considered valid, it must meet all the acceptable criteria including accuracy and precision, as well as any other criteria specified by the analytical method used. Therefore, all data points critical to the sampling program in terms of completeness will be 100% validated by the ERG contract in accordance with the ERG's generic Support For NMOO SNMOC, UATMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan, Section 15. With 100% validation, the rationale for considering data points non-critical is not required.

10.0 Performance and Systems Audits

10.1 Assessments and Response Actions •

No performance audit of field operations is anticipated at this time. I f conducted, performance and systems audits will be in accordance with:

• U.S. EPA (Environmental Protection Agency) Region II . April 2000. SOP SCST-1, Standard Operating Procedure (S.O.P.) for Performing Oversight of CERCLA Field Operations. Revision 0. Division of Environmental Services and Assessment, Hazardous Waste Support Branch, Hazardous Waste Support Section, Edison, NJ.

11.0 PREVENTIVE MAINTENANCE

11.1 Instrument/Equipment Testing, Procedures & Scheduled Inspection and Maintenance Requirements

As previously stated, calibration and preventative maintenance of analytical laboratory equipment will follow procedures as specified in ERG's generic Support For NMOO

16

SNMOC, UA TMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan, Section 15, which can be found as Appendix B.

11.2 Inspection/Acceptance Requirements for Supplies and Consumables

Due to the nature of air sampling rinsate and trip blanks are not applicable. SUMMA™ canister quality control includes calibration of the canister itself, method blanks performed by the laboratory and laboratory control samples, also performed by the laboratory.

12.0 SPECIFIC ROUTINE PROCEDURES/MEASUREMENT PARAMETERS INVOLVED

12.1 Reconciliation with Data Used to Assess PARCC for Quality Objectives Measurement

Sample collection precision will be evaluated by collecting and analyzing a field duplicate sample. The field duplicate samples will be used to evaluate errors associated with sample heterogeneity, sampling methodology and analytical procedures. The analytical results from the field duplicate will provide data on the overall measurement precision. Precision will be reported as the relative percent difference (RPD) for two measurements. The acceptance criteria for the field duplicate samples are located in Table 2 on page 17.

Data will be generated through the collection of air samples in residential basements in the area of the White Swan Laundry and Cleaners Site. This data will be used to determine the location and concentration of contamination in the residents, the extent of contamination, evaluate potential health threats, and determine environmental impacts while identifying clean-up criteria.

TABLE 2: PRECISION AND ACCURACY

Sample Parameter/Fraction

Sample Matrix

Analytical Method

Method Detection

Limit 1

Quantitation Limit

Estimated Accuracy1

Accuracy Protocol

Estimated Precision1

Precision Protocol

Title III Clean Air Amendment List

Volatile Organic Compounds (VOCs)

Air TO-15 0.04 -0.26 ppbv

ppbv <or= 30% Non-RAS % difference and absolute % difference

Non-RAS

' The method detection limits were provided by ERG as per their generic Support For NMOO SNMOC, UA TMP, and PAMS Networks, Contract No. 68-D-99-007, Quality Assurance Project Plan. ,

13.0 CORRECTIVE ACTION

13.1* Assessments and Response Actions

Procedures are provided for project personnel to make changes, take corrective actions and document the process through Corrective Action Request Forms. Corrective action can occur during field activities, laboratory analysis, data validation, and data assessment.

17

Corrective action in the field may be necessary when the monitoring network design is changed. A change in the field includes: increasing the number or type of samples or analyses; changing sampling locations; and/or modifying sampling protocol. When this occurs, the project officer or project QA officer will identify any suspected technical or QA deficiencies and note them in the field logbook. The project QA officer will be responsible for assessing the suspected deficiency and determining the impact on the quality of the data. Development of the appropriate corrective action will be the responsibility of the OSC.

Data validation and data assessment corrective action will be in accordance with the U.S. EPA Region II CERCLA Quality Assurance Manual.

14.0 QA REPORTS TO MANAGEMENT

14.1 Distribution List

The following project personnel will receive copies of the approved QAPP and any subsequent revisions.

Project Personnel Title

Tom Budroe On-Scene Coordinator ERRD/RAB

Diane Salkie Project Officer DESA/HWSB

Pat Sheridan Quality Assurance Officer DESA/HWSB

14.2 Reports to Management

The data collected as a result of sampling activities; will be organized, analyzed and summarized in a final project report that will be submitted to the OSC according to the Project Schedule. The report will be prepared by the project officer or project quality assurance officer and include appropriate data quality assessment.

APPENDIX A

SITE MAPS

I I_* ; ASBURY PARK, N.J. 40074-Bl-TF-024

Ai 1989 — A DMA6.64MNE-SERJ0SV822

Magnolia Avenue Ground Water Contamination Area MAP 2C

Outlined area depicts the extent of PCE, TCE and DCE contamination plume.

Coast Guard

1.6 Miles

Gulf Service Station Sun Cleaners White Swan L&C

Township of Wall Monmouth County. Tax Map 1969 Block 706 Lot 2

White Swan Laundry & Cleaners 1322 Sea Girt Avenue Wall Twp. Monmouth Co. New Jersey

MAP 2

S-1

GW-29

Parking S-2

GWl

GW-30 GW-2

GW-7

former location

of White Swan building

White Swan Launder & Cleaners

1322 Sea Girt Avenue

Wal Twp. Monmouth Co.

New Jersey

MAP 3

GW-1

C

Parking'

grass island 3

Of ass area

LEGEND • Ground Water Sample • Soil Sample A Ground Water and

Soil Sample

Gulf Service Station 1324 Sea Girt Ave. Wall Twp Monmouth County New Jersey .

MAP 3A

Sun Cleaners Site Map/Sample Location Map

MAP-4

Atlantic Ave

Sun Cleaners Building Sun Cleaners Property

200 200 400 600 800 Feet

APPENDIX B

Eastern Research Group (ERG)

Support for NMOC/SNMOC, UATMP and PAMS Networks, Contract No. 68-D-99-007. Quality Assurance Project Plan.

Morrisville, North Carolina

March 2000

SUPPORT FOR NMOC/SNMOC, UATMP, AND PAMS NETWORKS

Contract No. 68-D-99-007

2000

Quality Assurance Project Plan

Eastern Research Group, Inc. P.O. Box 2010

Morrisville, North Carolina 27560

Approved by:

ERG Program Manager:

ERG Deputy Program Manager: :

ERG Program QA Officer:

U.S. EPA Project Officer:

Project No. 0121.00 Element No. A2 Revision No. 1 Date March 2000 Page ii of viii

DISCLAIMER

This Quality Assurance Project Plan has been prepared specifically for the operation and management of the U.S. EPA National NMOC/SNMOC, UATMP, PAMS, and HAPS Programs. The contents have been prepared in accordance with Level III Specification of the EPA Guidance for Quality Assurance Project Plans, EPA QA/G-5.

Project No. Element No. Revision No Date Page

TABLE OF CONTENTS

Section

List of Tables : vi List of Figures vii Symbols and Abbreviations viii A PROJECT MANAGEMENT . . 1 of 6

1 Project/Task Organization 1 of 6 1.1 Assignment of Program Personnel 1 of 6

2 Problem Definition/Background 1 of 2 3 Project/Task Description and Schedule 1 of 15

3.1 NMOC and SNMOC . . l o f l 5 3.2 UATMP 6 of 15 3.3 PAMS . .' 10 of 15 3.4 HAPs 13 of 15

4 Data Quality Objectives and Criteria for Measurement Data l o f l 5 5 Special Training Requirements/Certification 1 of 2

5.1 Sampling Personnel •••• 1 of2 5.2 Analytical Laboratory Personnel 1 of 2

B MEASUREMENT DATA ACQUISITION 1 of 8 6 Sampling Process Design . 1 of 8

6.1 NMOC and SNMOC Sampling 1 of 8 6.2 UATMP Sampling . . . . 3 of 8 6.3 PAMS Sampling 8 of 8 6.4 HAPs Sampling 8 of 8

7 Sample Handling and Custody Requirements 1 of 17 7.1 NMOC, SNMOC, and UATMP Sample Custody 1 of 17 7.2 Carbonyl Sample Custody 12 of 17 7.3 HAPs Sample Custody ." 14 of 17 7.4 Sampling Monitoring Data 17 of 17

8 Analytical Methods Requirements 1 of 17 8.1 Canister Cleanup System l o f l 7 8.2 NMOC Analytical Systems 4 of 17 8.3 SNMOC Analytical Systems 6 of 17 8.4 UATMP and Concurrent Analytical System 6 of 17 8.5 PAMS Analytical Systems 9 of 17 8.6 Semivolatile Analytical Systems 11 of 17 8.7 Ethylene Oxide by Gas Chromatograph Analytical Systems . . . . 13 of 17 8.8 Dioxin/Furan by High Resolution Mass Spectroscopy

Analytical Systems 16 of 17 8.9 Metals Using an Inductively Coupled Argon Plasma Mass

Spectroscopy Analytical System 17 of 17

0121.00 A2

1 March 2000

iii of viii

glp\D:\QAPPCOV.WPD

Project No. 0121.00 Element No. A2 Revision No. 1 Date March 2000 Page iv of viii

TABLE OF CONTENTS (Continued)

9 Quality Control Requirements l o f l 6 9.1 Sample Canister Cleanup Studies l o f l 6 9.2 Standard Traceability , l o f l 6 9.3 Accuracy and Acceptance 2 of 16 9.4 Precision 9 of 16 9.5 Completeness • 9 of 16 9.6 Representativeness . . . . .. 11 of 16 9.7 Comparability 11 of 16 9.8 Lowest Quantitation Limits 11 of 16

10 Instrument/Equipment Testing, Inspection, and Maintenance Requirements 1 of 2 10.1 NMOC 1 of 2 10.2 SNMOC, UATMP, and PAMS 1 of 2 10.3 PAMS 2 of 2 10.4 HAPS 2 of 2

11 Instrument Calibration and Frequency 1 of 11 11.1 NMOC Calibration . . . l o f l l 11.2 SNMOC Calibration 3 of 11 11.3- UATMP Calibration 4 of 11 11.4 PAMS Calibration 7 of 11 11.5 HAPS Calibration 7 of 11

12 Data Management l o f l C ASSESSMENT/OVERSIGHT 1 of 3

13 Assessments and Response Actions 1 of 3 13.1 QA Performance Audits 1 of 3 13.2 Performance Evaluation and System Audits 1 of 3

. 13.3 QA Reports 1 of 3 14 Reports to Management : 1 of 2

14.1 QA and QC Functions 1 of 2 D DATA VALIDATION AND USABILITY 1 of 9

15 Data Review, Validation, and Verification Requirements 1 of 9 15.1 NMOC/SNMOC Data Reduction, Validation, and Reporting 2 of 9 15.2 UATMP Data Reduction, Validation, and Reporting 4 of 9 15.3 PAMS Data Reduction, Validation, and Reporting 6 of 9 15.4 HAPS Data Reduction, Validation, and Reporting 7 of 9 15.5 Aerometric Information Retrieval System Air Quality Subsystem

(AIRS AQS) . . : 8 of 9 16 Reconciliation with Data Quality Objectives l o f l 17 References 1 of 3

APPENDICES

glp\D:\QAPPCOV.WPD

Project No. Element No. Revision No. Date Page

0121.00 A2

1 March 2000

v of viii

TABLE OF CONTENTS (Continued)

The following methods are included by reference. These methods are available through the U.S. EPA Bulletin Board —www.epa.gov/:

ttn/amtic/airtox.htlm—

EPA Compendium Method TO-12, "Method for the Determination of Non-Methane Organic Compounds (NMOC) in Ambient air Using Cryogenic Preconcentration and Direct Flame Ionization Detection (PDFID)"

EPA Compendium Method TO-14A, "Determination of Volatile Organic Compounds (VOCs) in Ambient Air Using Specially Prepared Canisters with Subsequent Analysis by Gas Chromatography"

EPA Compendium Method TO-15, "Determination of Volatile Compounds (VOCs) in Air Collected in Specially Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS)"

EPA Compendium Method TO-11 A, "Determination of Formaldehyde in Ambient Air Using Adsorbent Cartridge Followed by High Performance Liquid Chromatography (HPLC) [Active Sampling Methodology]"

EPA Compendium Method TO-9A, "Determination of Polychlorinated, Polybrominated and Brominate/Chlorinated Dibenzo-p-Dioxins and Dibenzofurans in Ambient Air"

sw-846/sw-846.htm—

Method 8290, "Polychlorinated Dibenzodioxins (PCDDs) and Polychlorinated Dibenzofurans (PCDFs) High-Resolution Gas Chromatography/High-Resolution Mass Spectrometry (HRGC/HRMS)"

glp\D:\QAPPCOV.WPD

Project No. 0121.00 Element No. A2 Revision No. 1 Date March 2000 Page vi of viii

LIST OF TABLES

Table

1-1 1999/2000 Program Organization 1 of 6 I - 2 QC Responsibilities and Review Functions 5 of 6

3-1 SNMOC Target Compounds 4 of 15 3-2 UATMP Target Compounds 8 of 15 3-3 PAMS VOC Target Compounds 12 of 15 3-4 Carbonyl Target Compounds 14 of 15 3- 5 Analysis of Hazardous Air Pollutants .' 15 of 15

4- 1 NMOC Data Quality Objectives 2 of 15 4-2 Summary of SNMOC Procedures 3 of 15 4-3 Air Toxics TO-15 QC Procedures . . 4 of 15 4-4 Carbonyl Data Quality Objectives 7 of 15 4-5 Quality Control Procedures for Analysis of Semivolatile Organic Samples

According to EPA Method 8270 9 of 15 4-6 Quality Control Parameters for Ethylene Oxide Analysis Performed According to

the Analytical Procedures of NIOSH Method 1614 12 of 15 4-7 Quality Control Parameters for Dioxin/Furan Analysis Performed According to

the Analytical Procedures of EPA Method 8290 13 of 15 4-8 Quality Control Parameters for Phosgene Performed According to the

Analytical Procedures of Compendium TO-6 14 of 15 4-9 Quality Control Measures for Metals Analysis According to Method 10-3.5 . . . . 15 of 15

8-1 UATMP GC/FID/MSD Operating Conditions 7 of 17 8- 2 Semivolatile Organic Compounds to be Analyzed by the Analytical Procedures of

Method 8270, with Estimated Method Detection Limits 14 of 17

9- 1 Decafluorotriphenylphosphine (DFTPP) Key Ions and Ion Abundance Criteria According to EPA Method 8270 5 of 16

9-2 Quality Control Measures for Metals Analysis 10 of 16 9-3 SNMOC Lowest Quantitation Limits 13 of 16 9-4 TO-15 Analyte Lowest Quantitation Limit (LQL) 15 of 16 9-5 Lowest Quantitation Limits, Underivatized Concentration (ppbv) ,.> 16 of 16

I I - i Analytical Equipment Calibration Requirements . . 9 of 11

14-1 QC Responsibilities and Review Functions . 2 of 2

glp\D:\QAPPCOV.WPD

Project No. 0121.00 Element No. A2 Revision No. 1 Date ' March 2000 Page vii of viii

LIST OF FIGURES

Figure

I - 1 Program Organization Chart . . 2 of 6

6-1 NMOC, SNMOC, and 3-Hour Air Toxics Sampling System Components 2 of 8 6-2 Carbonyl Sampling System 4 of 8 6-3 Sampling Assembly for the UATMP .'. 6 of 8 6- 4 Cross-Sectional View of the Ozone Scrubber Assembly 7 of 8

7- 1 Canister Sample Data Sheet 2 of 17 7-2 Sample Receipt Login Information 3 of 17 7-3 Canister Tag 4 of 17 7-4 NMOC Invalid Sample Form 5 of 17 7-5 NMOC Daily HP 5880 Calibration Form 7 of 17 7-6 Canister Cleanup Log • 8 of 17 7-7 UATMP Analysis Log 10 of 17 7-8 Carbonyl Field Data Sheet 13 of 17 7-9 Label for Sample Identification 15 of 17 7- 10 Corrective Action Report 16 of 17

8- 1 Canister Cleaning Apparatus : 2 of 17 8-2 Schematic of Analytical Systems for NMOC 5 of 17 8-3 Gas Chromatograph/Mass Spectrometer/FID System 8 of 17 8-4 HPLCSystem 10ofl7

I I - 1 Dynamic Flow Dilution Apparatus 6 of 11

glp\D:\QAPPCOV.WPD

Project No. Element No. Revision No Date Page

SYMBOLS AND ABBREVIATIONS

ug Micrograms

uL Microliters

AC Area Counts

AIRS AQS Aerometric Information Retrieval System Air Quality Subsystem

BFB 4-Bromofluorobenzene

cm Centimeter

DNPH 2,4-Dinitrophenylhydrazine

EPA EnvironmentalProtection Agency

ERG Eastern Research Group, Inc.

FID Flame Ionization Detector

GC Gas Chromatograph

GC/MSD Gas Chromatograph/Mass Selective Detector

Hg Mercury

HPLC High Performance Liquid Chromatography

ID Identification

KI Potassium Iodide

m Meter

MB Megabyte

MDL Method Detection Limit

min Minute

mL Milliliter

mm Millimeter

MS/MSD Method Spike/Method Spike Duplicate

NAAQS National Ambient Air Quality Standard

NERL National Exposure Research Laboratory

NIST National Institute of Standards and Technology

nm Nanometer

NMOC Nonmethane Organic Compound

OAQPS Office of Air Quality Planning and Standards

0121.00 . A2

1 March 2000

viii of viii

glp\D:\QAPPCOV.WPD

Project No. Element No. Revision No. Date

0121.00 A2

1 March 2000

ix of viii Page

PAMS

PDF ID

ppbC -

ppbv

ppmC

pprhv

psig

QA/QC

QAD

QAPP

RCA

RPD

RSD

RTP

SIP

SNMOC

TAD

UAM

UATMP

VOC

SYMBOLS AND ABBREVIATIONS (Continued)

Photochemical Assessment Monitoring Station

Preconcentration Direct Flame Ionization Detection

Parts per Billion as Carbon

Parts per Billion Volume

Parts per Million as Carbon

Parts per Million Volume

Pounds per Square Inch Gauge

Quality Assurance/Quality Control

Quality Assurance Division

Quality Assurance Project Plan

Recommendation for Corrective Action

Relative Percent Difference

Relative Standard Deviation

Research Triangle Park

State Implementation Plan

Speciated Nonmethane Organic Compound

Technical Assistance Document for Sampling and Analysis of Ozone Precursors

Urban Airshed Model

Urban Air Toxics Monitoring Program

Volatile Organic Compound

glp\D:\QAPPCOV.WPD

Project No. 0121.00 Element No. A4 Revision No. 1 Date March 2000 Page 1 of 6

A—PROJECT MANAGEMENT

SECTION 1

PROJECT/TASK ORGANIZATION

1.1 Assignment of Program Personnel

Table 1-1 presents the 1999/2000 program organization listing the program assignment

and responsible person.

Table 1-1

1999/2000 Program Organization

Program Assignment Program Personnel Assigned

Program Manager Dave-Paul Dayton

Deputy Program Manager ; Julie Swift

Task Leader - Site Coordination/NMOC Analysis Peer Reviewer

Mitch Howell Julie Swift

Task Leader - SNMOC Analysis Peer Reviewer

Donna Tedder Julie Swift

Task Leader - Air Toxic Analysis Peer Reviewer

Mitch Howell Julie Swift

Task Leader - Carbonyl Analysis Peer Reviewer

Randy Bower Donna Tedder

Task Leader - PAMS Support Peer Reviewer

Dave-Paul Dayton Rob Martz

Task Leader - HAPS Support Peer Reviewer

Rob Martz Julie Swift

Task Leader - Reporting Peer Reviewer

Julie Swift Dave-Paul Dayton

Task Leader - AIRS Peer Reviewer

Randy Bower Dave-Paul Dayton

Program QA Officer Joan Bursey

Senior Technical Advisor Ray Merrill

Project Administrator Carol Hobson

Project Secretary Gail Pierce

The program organizational chart is presented in Figure 1-1.

glp/D:\SECT1.WPD

" D

o

o

o EPA Project Officer

Vickie Presnell

EPA Delivery Order Manager Sharon Nizich

SNMOC Analysis Task Leader

Donna Tedder

Peer Reviewer Julie Swift AirT

Task Mitch

oxics Leader Howell

Peer Reviewer Julie Swift

Program QA Officer Joan Bursey

Program Manager Dave-Paul Dayton

Senior Technical Advisor

Raymond Merrill Deputy Program Manager Julie Swift

Project Administrator Carol Hobson

Project Secretary Gail Pierce

1

Carbonyis Analysis Task Leader

Randy Bower

Peer Reviewer Donna Tedder

NMOC Analysis & Site Coordination

Task Leader Mitch Howell

PAMS Support Task Leader

Dave-Paul Dayton

Peer Reviewer Julie Swift

HAPS Support Task Leader Rob Martz

Peer Reviewer Rob Martz

AIRS Dz TaskL

Randy

tabase eader 3ower

Peer Reviewer Dave-Paul Dayton

Peer Reviewer Julie Swift Rep

TaskL Julie S

Drt =ader swift

Peer Reviewer Dave-Paul Dayton

y o 5* P m o °5 I < C/5

o '

o

CD " ,

CD' o

O 2. o

3 o

M to N> Figure 1-1. Program Organization Chart o ^ ^ 0\ O H o

Project No. Element No. Revision No. Date

0121.00 A4

1 March 2000

3 of 6 Page

1.1.1 Program Manager

ERG's program manager, Dave-Paul Dayton, a member of ERG's technical staff, has the

primary concern of understanding the state's and EPA's needs at the program level and ensuring

overall timely performance of high quality technical services. He coordinates with the technical

advisors, peer reviewers, coordinators, directors, and the task leaders to communicate technical

issues and needs and to ensure that these individuals are involved in management decisions

appropriate for their roles in this contract.

1.1.2 Deputy Program Manager

As the Deputy Program Manager, Julie Swift is responsible on a day-to-day basis for the

technical conduct of the program and for leading the analytical tasks and providing technical

direction and support. She assists with any technical problems that arise. She responds to the

task leaders regarding any project issues that affect their task(s). She assists with analytical

analysis, data reduction, review and reporting. She is responsible for ensuring that the

appropriate level of staffing, number of work shifts, and committed resources (automated

analytical equipment) exist to meet the required project deliverables and sample turnaround time.

She tracks budget performance for all tasks and reports this information to the Program Manager

and all Task Leaders. She also ensures that all management systems and tools required for this

program are implemented and tracked, and tracks deliverables and budget performance to present

project performance information to the EPA at monthly meetings and in monthly progress

reports.

1.1.3 Program QA Officer

The Program QA Officer, Joan Bursey, is responsible for ensuring the overall integrity

and quality of the project results. She reviews the QAPP and coordinates data and laboratory

audits that will provide information relative to data quality and determine whether procedures are

glp/D:\SECT1.WPD

Project No. 0121.00 • Element No. A4

Revision No. 1 Date March 2000 Page 4 of 6

in accordance with the QAPP. The lines of communication between management, the Program

QA Officer, and the technical staff are formally established and allow for discussion of real and

potential problems, preventive actions, and corrective procedures. The major QC responsibilities

and QC review functions are summarized in Table 1-2.

Anytime during the program, additional QA/QC measures may be initiated upon

consultation between the task leader, Program QA Officer and the senior technical advisor.

1.1.4 Senior Technical Advisor

The senior technical advisor, Dr. Raymond G. Merrill, is responsible for ensuring the

overall technical quality of ERG's approach to the program. Dr. Merrill's ultimate responsibility

is ensuring client satisfaction and that components of effective management are active at all

times during the contract performance period.

1.1.5 Task Leaders

The ERG task leaders are responsible for meeting the project objectives, meeting budgets

and schedules, and directing the technical staff in execution of the technical effort for their

respective task(s). The task leaders manage the day-to-day technical activities. They assess and

report on the project's progress and results (e.g., recordkeeping, data validation procedures,

sample turnaround time), and ensure timely, high-quality services and adherence to the project

QA plan.

glp/D:\SECT1.WPD

Project No. 0121.00 Element No. A4 Revision No. 1 Date March 2000 Page 5 of 6

Table 1-2

QC Responsibilities and Review Functions

Responsible Person Major Responsibilities

Program Manager • Ensure overall timely performance of high quality technical services • Communicate technical issues and needs • Track all management systems and tools • Track deliverables and budget performance • Review reports before reporting to the client

Deputy Program Manager

• Ensure data quality • Check information completeness • Assist with technical problems • Ensure appropriate leyel of staffing and committed resources exist to perform

work • Review data completeness and quality before reporting to client • Review all reports • Report project performance (budget and deliverables) to EPA at monthly

meetings and in monthly progress reports • Day-to-day management of task leaders

Program QA Officer Review QC reports • Make QA recommendations • Write and/or review test plan • Write and/or review QAPP • Audit laboratory(s) • Review documentation (reports, etc.)

Technical Advisor • Propose procedural change • Propose equipment change • Assist with technical problems.

Peer Reviewer • Ensure final data quality • Final data review • Assist with technical problems

Analytical Task Leader • Review documentation • Develop analytical procedures • Propose procedural changes .

Data review and validation • Analyst training and supervision • Meet task budgets and report schedules • Manage day-to-day technical activities • Check information completeness • Review instrument and maintenance log books • Review calibration factor drift • Perform preventive maintenance • Prepare monthly/quarterly reports

1.1.6 Peer Reviewers

glp/D:\SECT1.WPD

Project No. Element No. Revision No. Date

0121.00 A4

1 March. 2000

6 of 6 Page

The ERG peer reviewers are responsible for ensuring the final data quality before

anything is reported to the client. They perform the final data review on the analytical reports.

The peer reviewers also assist in resolving any technical problems that occur in the laboratory or

at the sites.

glp/D:\SECT1.WPD

Project No. 0121.00 Element No. A5 Revision No. 1 Date March 2000 Page 1 of2

SECTION 2

PROBLEM DEFINITION/BACKGROUND

The Clean Air Act Amendments of 1990 required the Environmental Protection Agency's

(EPA's) Office of Air Quality Planning and Standards (OAQPS) to set National Ambient Air

Quality Standard (NAAQS) for the "criteria" pollutant, ozone. In areas of the country where the

NAAQS for ozone is being exceeded, additional measurements of the ambient nonmethane

organic compound (NMOC)(1) concentration are needed to assist the affected states in developing

revised ozone control strategies. Measurements of ambient NMOC are important to the control

of volatile organic compounds (VOCs) that are precursors to atmospheric ozone. Because of

previous difficulty in obtaining accurate NMOC concentration measurements, EPA and Radian

Corporation started a monitoring and analytical program in 1984 to provide support to the states.

Studies indicate that a potential for elevated cancer risk is associated with certain toxic

compounds often found in urban ambient air.(2) In 1987, EPA developed the Urban Air Toxics

Monitoring Program (UATMP) to help State arid local agencies characterize the nature and

extent of potentially toxic air pollution in urban areas. Since 1987, several State and local

agencies have participated in the UATMP by implementing ambient air monitoring programs.

These efforts have helped to identify the toxic compounds most prevalent in the ambient air and

indicate emissions sources that are likely to be contributing to elevated concentrations. As a

screening program the UATMP also provides data input for models used by EPA (and others) to

assess risks posed by the presence of toxic compounds in urban areas. The UATMP program is a

year-round sampling program, collecting 24-hour integrated ambient air samples at urban sites in

the contiguous United States every 12 days, and is also supported by ERG.

The speciated NMOC (SNMOC) program was initiated in 1991 in response to requests

by State agencies for more detailed speciated hydrocarbon data for use in ozone control strategies

and Urban Airshed Model (UAM) input. In 1996, Radian Corporation sold the EPA contracts

glp/D:\SECT2.WPD

Project No. Element No. Revision No. Date

0121.00 A5

1 March 2000

2 of 2 Page

and necessary resources to Eastern Research Group, Inc. (ERG), who now support EPA for the

NMOC and SNMOC programs.

Title I , section 182 of the Clean Air Act Amendments of 1990 requires states to establish

Photochemical Assessment Monitoring Stations (PAMS) as part of their state implementation

plan (SIP) for ozone nonattainment areas. The rule revises the ambient air quality surveillance

regulations to include enhanced monitoring of ozone and its precursors. The regulations

promulgated in 1993 require monitoring of ozone, oxides of nitrogen (NOx), selected carbonyl

compounds, and VOCs. The required monitoring is complicated and requires considerable lead

time for the agencies to acquire the equipment and expertise to implement their PAMS network.

Under the PAMS program, each site may require a different level of support with respect to

sampling frequency, sampling equipment, analyses, and report preparation. Presampling,

sampling, and analytical activities are performed according to the guidance provided in the

Technical Assistance Document for Sampling and Analysis of Ozone Precursors (TAD), 1998

revision/3' The specific methodology applicable to the PAMS program will be discussed in this

Quality Assurance Project Plan (QAPP).

In 1999, the EPA expanded this program to provide for the measurement of additional

Clean Air Act Hazardous Air Pollutants (HAPs) to support the Government Performance and

Results Act (GPRA). As required under the GPRA, the EPA developed a Strategic Plan that

includes a goal for Clean Air. Under this goal, there is an objective to improve air quality and

reduce air toxic emissions to levels 75 percent below 1993 levels by 2010 in order to reduce the

risk to Americans of cancer and other serious adverse health affects caused by airborne toxics.

This combined QAPP defines the presampling and sampling activities and laboratory

analyses conducted by ERG for the NMOC, SNMOC, UATMP, PAMS, and HAPs programs and

describes the quality assurance/quality control (QA/QC) procedures used to assess data quality.

Many of these procedures are based on previous NMOC studies.(4"'0,12"l6>

glp/D:\SECT2.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

1 of 15

SECTION 3

PROJECT/TASK DESCRIPTION AND SCHEDULE

This section describes the acti vities performed under each of the major program

components (NMOC, SNMOC, UATMP, PAMS, and HAPs). The SUMMA® canisters used by

this laboratory are dedicated to each separate program. Sampling and analysis schedules are

prepared in the project instructions when the delivery orders are provided by EPA.

3.1 NMOC and SNMOC

The NMOC and SNMOC programs require several activities for a successful monitoring

program. The monitoring program begins with presample collection activities. The NMOC and

SNMOC sample collection systems are designed to collect ambient air samples in

SUMMA®"treated stainless steel canisters over a 3-hour period. The sample collection period

occurs from 6:00 - 9:00 a.m. local time to capture mobile source pollutants during the morning

"rush hour" simultaneously with sunrise, which provides the energy necessary for many

photochemical reactions.

A selected number of canisters from state and EPA directed sites are analyzed for

additional air toxic compounds; the sites and canisters are identified at the beginning of the

program to ensure sample completeness. Some sites also collect carbonyl samples for analysis..

The analytical methods and procedures are discussed later in the UATMP and PAMS project

descriptions.

The SUMMA® canisters dedicated to the program are checked for leaks, repaired, and

cleaned using a vacuum and pressurization canister cleaning system. The canisters are certified

by ERG for cleanliness by analyzing the contents using EPA Compendium Method TO-12 for

determining total NMOC concentration.

glp/D:\SECT3.WPD

Project No. 0121.00 Element No. A6 Revision 1 Date March 2000 Page 2 of 15

The State or local agency site personnel are contacted to coordinate site installation,

operator training, sample collection, and shipping. ERG provides installation of the sample

collection system, supporting documentation, training of the site operator for collection of

scheduled samples, and ongoing technical support and coordination for sample collection during

the entire monitoring program.

Samples are collected by State or local agency personnel every weekday starting on the

first Monday of June through the end of September at each of the designated sites. At least two

days before each sample collection episode, ERG ships the necessary clean, certified canisters to

the site along with the field sample collection form and chain of custody forms. The time

integrated ambient samples are then collected and shipped to ERG for analysis.

Samples are delivered to a dedicated loading dock area that is part of the laboratory space

used for the programs. Samples are received and logged into a sample receipt log and into a

computerized login database networked to be accessible to all analysts and task leaders. After

the sample identification number, date received, sample date, project name, canister pressure, and

storage location are documented, the field sample collection form is reviewed and any

discrepancies or invalidated samples are reported to the Deputy Program Manager. ERG

contacts the site operator for resolution of any sample issues. The samples are then taken to the

laboratory for analysis.

The analytical equipment used for the NMOC program consists of two modified

Preconcentration Direct Flame Ionization Detection (PDFID) Hewlett-Packard gas

chromatographs (GC) with cryogenic sample preconcentration systems and dual-channel Flame

Ionization Detectors (FIDs). EPA Compendium Method TO-12 is used for the analysis.

The PDFID systems used for analyses are calibrated and blanked daily before sample

analysis. Cleaned, humidified air from the canister cleaning system is analyzed to determine the

level of organic compounds present in the analytical system. Upon achievement of acceptable

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

3 of 15

system blank results (< 10 ppbC), a daily QC check sample of propane is analyzed. The

QC check sample is used to check the calibration of the analytical system. Upon acceptable

calibration results (r2 2:0.995), sample analysis begins.. Ten percent of the total number of

samples received are collected in duplicate and analyzed twice to determine the precision and

analysis for the program.

The NMOC data are then processed to determine the total NMOC present in the sample.

The parts per million as carbon (ppmC) concentration of the NMOC is determined using the

daily propane calibration response factor. Preliminary data summaries are compiled monthly for

all sites and distributed to the site contacts and the EPA Project Officer.

During the 1997 season, ERG's laboratory implemented a system on the standard

UATMP instrumentation to analyze the SNMOC canisters. For the first time, all analyses -

SNMOC, UATMP and PAMS compounds. - can be obtained from one analysis. Because of this

analytical achievement, effort and costs for any combination of analyses are significantly

reduced. ,

Speciated NMOC analysis is performed to identify and quantify the VOC species present

in the ambient air. The analytical equipment used,for the SNMOC program consists of an

Entech 7100 Preconcentrator, a Hewlett-Packard GC/FID/MSD, and a data acquisition system.

ERG staff analyze the samples for SNMOC compounds (listed in Table 3-1) in accordance with

the methodology specified in the TAD ( 3 ) using a GC/MSD and an FID following EPA

Compendium Methods TO-14A and TO-15. The FID is used to perform quantitative analysis of

the SNMOC compounds of interest; the MSD is used for confirmation and identification of .

glp/D:\SECT3.WPD

Project No. 0121.00 Element No. A6 Revision 1 Date March 2000 Page 4 of 15

Table 3-1

SNMOC Target Compounds

Compound

Ethylene 2,3-Dimethylpentane Acetylene 3-Methylhexane Ethane 1-Heptene Propylene 2,2,4-Trimethylpentane Propane «-Heptane Propyne : Methylcyclohexane Isobutane 2,2,3-Trimethylpentane Isobutene 2,3,4-Trimethylpentane 1 -Butene Toluene 1,3-Butadiene 2-Methylheptane «-Butane 3-Methylheptane /ra;«-2-Butene 1-Octene cw-2-Butene n-Octane 3-Methyl-l-Butene Ethylbenzene Isopentane p,w-Xylene 1-Pentene Styrene 2-Methyl-1 -Butene o-Xylene «-Pentane 1-Nonene Isoprene n-Nonane

. ;ra«.y-2-Pentene Isopropylbenzene c';i-2-Pentene «-Propylbenzene 2-Methyl-2-Butene • a-Pinene 2,2-Dimethylbutane (Neohexane) • m-Ethyltoluene Cyclopentene /7-Ethyltoluene 4-Methyl-l-Pentene 1,3,5 -Trime thylbenzene 2,3-Dimethylbutane o-Ethyltoluene Cyclopentane P-Pinene 2-Methylpentane (Isohexane) 1,2,4-Trimethylbenzene 3-Methylpentane 1-Decene 2-Methyl-1-Pentene «-Decane 1-Hexene 1,2,3 -Trime thylbenzene 2-Ethyl-l-Butene w-Diethylbenzene «-Hexane p-Diethylbenzene trans-2-Hexene 1 -Undecene cw-2-Hexene n-Undecane Methylcyclopentane Dodecene 2,4-Dimethylpentane n-Dodecane Benzene Tridecene Cyclohexane n-Tridecane 2-Methylhexane (Isoheptane) Total NMOC

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

5 of 15

compounds of interest. The FID provides good sensitivity and uniform response based on the

number of carbon atoms per compound.

Moisture and carbon dioxide are removed from the analytical system using a microscale

purge and trap dehydration device located in the Entech 7100 Preconcentrator. Personnel

perform cryogenic concentration of the samples using a trap consisting of chromatographic-grade

stainless steel tubing packed with commercially available hybrid 60/80 mesh Tenax®/deactivated

glass beads maintained at -160°C during sample concentration. The concentrated VOCs are

thermally desorbed at room temperature to revolatilize them for transfer to the secondary trap.

The secondary trap is Tenax® at -60°C. The VOCs are then back-flushed while heating to be

further focused on an open-tubular focusing trap for rapid injection onto the analytical column.

The sample is injected onto the cold column to separate C2 through C1 3 hydrocarbons and to

obtain a total SNMOC concentration.

The SNMOC systems are calibrated monthly using propane and blanked daily prior to

sample analysis. A QC standard is analyzed daily prior to sample analysis to ensure the validity

of the current monthly response factor. Following the daily QC standard analysis, cleaned, dried

air that has been humidified from the canister cleaning system is analyzed to determine the level

of organic compounds present in the analytical system. Upon achieving acceptable system blank

results, sample analysis begins. Samples are analyzed for the target compounds listed in

Table 3-1. Ten percent of the total number of samples are analyzed twice to determine the

precision of analysis for the program.

The SNMOC raw data from the PE-Turbochrom® (Perkin Elmer) chromatography data

acquisition system are processed and reduced to determine peak identifications for any target

analytes present in the samples. The propane response factor from the calibration curve

determines the parts per billion as carbon (ppbC) concentration of the target analytes.

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

6 of 15

At the end of the sample collection period, the postsampling activities begin. The sample

collection equipment is recovered from the sites and refurbished as necessary by ERG, who

collect and store the equipment in a dedicated area until the next monitoring program

presampling activities begin. ERG then prepares the final program report describing procedures,

results, discussion of results, compilation of statistics, and recommendations. Upon approval by

the EPA Project Officer and Delivery Order Manager, ERG distributes the final report to

designated persons. ERG provides the final data summaries to EPA in Excel® format on

magnetic floppy disk media, archives all project files, raw data, reports, correspondence, memos,

letters, and copies of the final report, and formats the finalized data for input into the AIRS AQS

3.2 UATMP

The UATMP requires several key activities for a successful monitoring program. The

program originates with presample collection activities. The UATMP sample collection system

is designed to collect whole-air 24-hour integrated ambient air samples in SUMMA®-treated

stainless steel canisters, resulting in a subatmospheric final pressure. Prior to field installation,

the sample collection systems are certified using a dual-manifold certification system, which

verifies cleanliness and determines the background level of target organic compounds introduced

by the sample collection system. The certification procedure also determines the percent

recovery of selected target analytes by challenging the system with a known concentration of

selected toxic organic compounds.

The SUMMA® canisters are checked for leaks, repaired i f necessary by ERG or the

canister vendor, cleaned using a vacuum and pressurization canister cleaning system, and then

certified for cleanliness. The cleanliness of a canister is determined by analyzing the contents

using EPA Compendium Method TO-15 for determining volatile compound concentration and

by analyzing one canister per cleaned set by Gas Chromatograph/Flame Ionization

Detector/Mass Selective Detector (GC/FID/MSD).

database.

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

7 of 15

The analytical equipment used for the UATMP consists of a cryogenic sample

concentration system and a GC/FID/MSD with a FID detector for hydrocarbon analysis. The

FID is used concurrently with the MSD to quantitate the 59 target compounds present in the

sample. UATMP target compounds are listed in Table 3-2. This system provides the required

sensitivity and confirmation of target compound identification to determine the detection limits

needed for the assessment of potential risks associated with the toxic compounds measured for

this program. The EPA Compendium Method TO-14A is followed as well as TO-15 to illustrate

that analyses for all compounds requested can be achieved depending on EPA's preference for

method.

As with NMOC and SNMOC activities, the State or local agency site personnel are

contacted to coordinate installation, operator training, sample collection, and shipping activities.

ERG provides installation of the sample collection system, support documentation, training of

the site operator for collection of scheduled samples, and ongoing technical support and

coordination of sample collection.

Samples are collected by State or local agency personnel once every 12 days for a period

of 1 year at each of the designated sites. At least 2 days prior to the sample collection episode,

ERG ships the necessary cleaned and certified canisters to the site along with the chain of

custody form and field sample collection form. The ambient air samples are collected in

canisters over a 24-hour period from midnight to midnight local standard time. Ten percent of

the total number of samples are received in duplicate and analyzed in replicate to statistically

determine the precision of sampling and analysis for the program.

glp/D:\SECT3.WPD

Project No. 0121.00 Element No. A6 Revision 1 Date March 2000 Page 8 of 15

Table 3-2

UATMP Target Compounds

UATMP Target Compounds

Acetylene Ethyl Acrylate

Propylene Bromodichloromethane

Dichlorodifluoromethane Trichloroethylene

Chloromethane Methyl Methacrylate

Dichlorotetrafluoroethane cis-1,3 - D ichloropropene

Vinyl Chloride Methyl Isobutyl Ketone

1,3-Butadiene ira^-l^-Dichloropropene

Bromomethane 1,1,2-Trichloroethane

Chloroethane Toluene

Acetonitrile Dibromochloromethane

Trichlorofluoromethane 1,2-Dibromoethane l

Acrylonitrile t

n-Octane 1,1-Dichloroethene Tetrachloroethylene

Methylene Chloride Chlorobenzene

Trichlorotrifluoroethane Ethylbenzene

trans-1,2-Dichloroethylene m-//?-Xylene

1,1-Dichloroethane Bromoform

Methyl te/Y-Butyl Ether Styrene

Methyl Ethyl Ketone 1,1,2,2-Tetrachloroethane

Chloroprene o-Xylene

cis-1,2-Dichloroethene 1,3,5-Trimethylbenzene

Bromochloromethane 1,2,4-Trimethylbenzene

Chloroform m-Dichlorobenzene

Ethyl te>7-Butyl Ether. Chloromethylbenzene

1,2-Dichloroethane /7-Dichlorobenzene

1,1,1 -Trichloroethane '• o-Dichlorobenzene

Benzene 1,2,4-Trichlorobenzene

Carbon Tetrachloride Hexachloro-1,3-Butadiene

fcrt-Amvl Methvl Etlier 1,2-DichloroDrorjane

Samples are shipped to ERG and received in a loading dock area that is part of the

dedicated laboratory space used for the program. ERG then logs the samples into the sample

glp/D:\SECT3.WPD

Project No. 0121.00 Element No. A6 Revision 1 Date March 2000 Page 9 of 15

receipt log book and documents the sample identification number, date received, sample date,

project name, canister vacuum, and storage location. ERG also logs samples into the •

computerized login database. After comparing the above information with the field sample

collection form, ERG brings any discrepancies or invalidated samples to the attention of the

Deputy Program Manager. ERG contacts the site operator for resolution of any sample issues,

and the samples are then taken to the laboratory for analysis.

The GC/FID/MSD system is calibrated for the target compounds in Tables 3-1 and 3-2

and blanked daily prior to sample analysis. The validity of the tune of the MSD is verified daily

using 4-Bromofluorobenzene (BFB). A QC check sample is also analyzed daily using a

UATMP standard and PAMS standard to validate the response factors from the calibration of the

analytical system. Upon acceptable QC results, a daily blank sample is analyzed. Clean,

humidified air from the canister cleaning system is analyzed to determine the level of organic

compounds present in the analytical system; upon acceptable blank results, sample analysis

begins.

ERG uses Hewlett-Packard Chemstation® and Perkin Elmer Turbochrom® data systems to

acquire data. Personnel identify compounds by referring to a combination of the compound's

retention time, the MSD library, and the analyst's experience and judgment. All of the target

UATMP compounds are quantitated using the MSD; all target SNMOC compounds are

quantitated using the FID. Sample concentrations are calculated using the monthly calibration

curve response factor from the MSD and propane monthly calibration for the FID. Preliminary

data summary reports are compiled every quarter for all sites and distributed to the site contacts

and the EPA Project Officer. ERG staff also finalize and format data for input into the AIRS

AQS database.

ERG oversees recertification and refurbishment of the samplers once a year to enable

sampling to continue from season to season without interruption. Staff prepare the final program

report describing the procedures, results, discussion of results, compilation of statistics, and

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

lOof 15

recommendations and, upon approval of the report by the EPA Project Officer, distribute the

final report to the sites and other persons as designated by the Project Officer. ERG staff also

provide the final data summaries to the EPA in Excel® format on magnetic floppy disk media.

All project files, raw data, reports, correspondence, memoranda, letters, and copies of the final

report are put in short-term file storage and archived.

3.3 PAMS

The program objective of PAMS is to provide data that are consistent with the proposed

rule for Ambient Air Quality Surveillance in accordance with 40 CFR Part 58. As a team, the

ERG staff can offer site support to any state that needs to set up a PAMS site or maintain it with

technical help.

After a PAMS program has been established by the State or local agency, ERG contacts

the EPA site personnel to coordinate sample collection and sample shipment. ERG maintains

coordination of the sample collection and sample shipments with the site contact and resolves

any issues that occur during the sampling season.

The State or local agency typically provides the program's SUMMA®-treated canisters.

ERG cleans the canisters using a vacuum and pressurization canister cleaning system, and then

certifies them for cleanliness by analyzing the contents using EPA Compendium Method TO-12

for determining total NMOC. Canisters are recycled through the canister cleaning and

verification process as needed to support the sample collection schedule for the program.

Sep-Pak® chromatographic-grade silica gel cartridges are used for carbonyl sample

collection. The vendor precoats the cartridges with 2,4-Dinitrophenylhydrazine (DNPH). A

potassium iodide (KI) ozone scrubber, actively maintained at about 65°C during sample

collection, is required to remove ozone from the sample stream.

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

11 of 15

Following sample collection, the site contact ships the canisters, cartridges, and

documentation to the ERG laboratory. Project personnel receive samples in a loading dock area

that is part of the dedicated laboratory space used for the programs, log them into the sample

receipt log book and the computerized log, and document the information pertaining to the

sample identification number, the date received, sample date, project name, canister vacuum, and

storage location. Project personnel review the chain of custody and field sample collection forms

and any discrepancies or invalidated samples are brought to the attention of the PAMS Task

Leader. If necessary, the Site Preparation Task Leader contacts the site for resolution of issues

for subsequent samples. The canister samples are then taken to the laboratory for analysis and

the cartridges are stored under refrigeration.

ERG staff analyze samples for PAMS VOC (listed in Table 3-3) in accordance with the

methodology specified in the TAD ( 3 ) using a GC/MSD and a FID. The FID is used to perform

quantitative analysis of the compounds of interest; the MSD is used for confirmation and

identification of compounds of interest. The FID provides good sensitivity and uniform response

based on the number of carbon atoms per compound. Moisture is removed from the FID

analytical system using a microscale purge and trap dehydration device. Personnel perform

cryogenic concentration of the samples using a trap consisting of chromatographic-grade

stainless steel tubing packed with commercially available 60/80 mesh deactivated glass beads

maintained at -180°C during sample concentration. The concentrated VOCs are thermally

desorbed at room temperature to revolatilize them for transfer to the secondary trap. The second

trap is Tenax® at -60°C. The VOCs are then back-flushed while heating to be further focused on

an open-tubular focusing trap for rapid injection onto the analytical column.

glp/D:\SECT3.WPD

Project No. Element No Revision Date • Page

Table 3-3

PAMS VOC Target Compounds

, Compound

Acetylene 3-Methylhexane Ethylene 2,2,4-Tri methylpentane Ethane n-Heptane Propylene Methyl cycl ohexane Propane 2,3,4-Trimethylpentane

. Isobutane Toluene 1-Butene 2-Methylheptane n-Butane 3-Methylheptane trans-2-Buterie «-Octane m-2-Butene Ethylbenzene Isopentane m-Xylene 1-Pentene ^-Xylene rc-Pentane Styrene Isoprene o-Xylene ;ra«s-2-Pentene «-Nonane c/s-2-Pentene Isopropylbenzene 2,2-Dimethylbutane ^-Propylbenzene Cyclopentane m-Ethyltoluene 2,3-Dimethylbutane p-Ethyltoluene 2-Methylpentane 1,3,5-Trimethylbenzene 3-Methylpentane o-Ethyltoluene 1-Hexene 1,2,4-Trimethylbenzene n-Hexane «-Decane Methylcyclopentane 1,2,3-Tri methy lbenzene 2,4-Dimethylpentane m-Diethylbenzene Benzene /j-Diethylbenzene Cyclohexane «-Undecane 2-Methylhexane Total NMOC 2,3-Dimethylpentane

0121.00 A6

1 March 2000

12 of 15

glp/D:\SECT3.WPD

Project No. 0121.00 Element No. A6 Revision 1 Date March 2000 Page 13 of 15

The carbonyl samples are analyzed for the carbonyl compounds listed in Table 3-4 using

EPA Compendium Method TO-11 A. The analytical instrument consists of a Varian 5000 High

Performance Liquid Chromatograph (HPLC) with a multiwavelength UV detector operated at

360 nanometers (nra). The HPLC is configured with a 25 centimeter (cm), 4.6 millimeter (mm)

ID CI 8 silica analytical column with a 5-micron particle size. Typically, 25 microliter (uL)

aliquots are injected with an automatic sample injector.

A PE-Turbochrom® chromatography data acquisition system is used to retrieve data from

both the ozone precursor and carbonyl analytical instruments. The data are processed and peak

identifications are made using retention times and relative retention times. After peak

identifications are made, the concentration of each target analyte is determined using individual

response factors for carbonyl compounds or propane response factors for ozone precursor

compounds. Preliminary data summary reports are distributed to the sites and the EPA Project

Officer once per month. Final data summary and a letter report are provided to the sites and the

EPA at the program end. Final data summary information is formatted for inclusion into the

AIRS AQS database upon approval by the EPA Project Officer.

3.4 HAPs

The program objective of HAPs is to provide data that are needed to support the

Government Performance and Results Act (GPRA). As a team, and with assistance from

Contractors, the ERG staff can offer site support to any state that needs HAPs analysis. The

responsibility for the equipment for sample collection falls on the state or local agency. The

analytical services support for this line item is shown in Table 3-5.

glp/D:\SECT3.WPD

Project No. Element No Revision Date Page

Table 3-4

Carbonyl Target Compounds

Compounds

Formaldehyde Isovaleraldehyde Acetaldehyde Valeraldehydes Propionaldehyde Tolualdehydes Crotonaldehyde Hexaldehyde Butyraldehyde 2,5-Dimethylbenzal dehyde Isobutyraldehyde Acetone Benzaldehyde

0121.00 A6

1 March 2000

14 of 15

glp/D:\SECT3.WPD

Project No. Element No. Revision Date Page

0121.00 A6

1 March 2000

15 of 15

Table 3-5

Analysis of Hazardous Air Pollutants

Analytical Analytical HAP Method HAP Method

Category I Category IV

Benzene TO-14A/TO-15 Acenaphthene TO-13A Carbon Tetrachloride TO-14A/TO-15 Acenaphthylene TO-13A Chloroform TO-14A/TO-15 Anthracene . TO-I3A Chloroprene TO-14A/TO-I5 Benzo(ghi)perylene TO-13A 1,4-Dichlorobenzene TO-14A/TO-15 Fluoranthene TO-13A Ethylene Dibromide TO-14A/TO-15 Fluorene TO-13A Ethylene Dichloride TO-14A/TO-15 Naphthalene TO-13A Hexachlorobenzene TO-14A/TO-15 Phenanthrene TO-13A Methyl Bromide TO-14A/TO-15 Pyrene TO-13A Methyl Chloride TO-14A/TO-15 Benz(a)anthracene TO-13A Styrene TO-14A/TO-15 Benzo(a)pyrene TO-13A Tetrachloroethylene TO-14A/TO-I5 Benzo(b)fluoranthene TO-13A Toluene TO-14A/TO-15 Benzo(k)fluoranthene TO-13A Trichloroethylene TO-14A/TO-15 Chrysene TO-13A Vinyl Chloride TO-14A/TO-15 Dibenz(a,h)anthracene TO-I3A Xylenes TO-14A/TO-I5 Indeno( 1,2,3-cd)pyrene TO-I3A 1,3-Butadiene TO-14A/TO-15 Acrylonitrile TO-14A/TO-15

Category II Category V

Acetaldehyde TO-11A Antimony & Compounds 10-3.5 Formaldehyde TO-11A Arsenic & Compounds 10-3.5

Beryllium & Compounds 10-3.5 Cadmium & Compounds 10-3.5 Chromium & Compounds* 10-3.5 Lead & Compounds 10-3.5 Manganese & Compounds 10-3.5 Mercury & Compo'unds IO-3.5 Nickel & Compounds 10-3.5

Category III

Phosgene TO-6 bis(2-Chloroethyl) Ether TO-13A bis(2-Ethylhexyl) Phthalate TO-13A 2,3,7,8-Tetrachlorodibenzo-p-Dioxin TO-9 Ethylene Oxide NIOSH 1614 ,.

*Chromium determined from a filter is total chromium, not chromium VI. Chromium VI oxidizes when sampled on a filter.

glp/D:\SECT3.WPD

Project No. Element No. Revision No Date Page

SECTION 4

DATA QUALITY OBJECTIVES AND CRITERIA FOR MEASUREMENT DATA

The data quality objectives for the four programs—NMOC, UATMP, PAMS, and

HAPs—are similar but are not identical. Therefore, the programs are discussed separately.

Data quality objectives are presented in Table 4-1. for the NMOC monitoring program

and in Table 4-2 for the SNMOC monitoring program. The major quality objectives of these

programs are to ensure that ambient air samples are collected in a prescribed manner and NMOC

and SNMOC concentrations are measured precisely and accurately. Because the SNMOC

samples are also analyzed by the UATMP system, the quality objectives presented in Table 4-3

are adhered to also when applicable to hydrocarbon analyses (flagged with an ®).

The quality objectives of the UATMP and the HAPs supported Category I Analytes listed

in Table 3-5 are to ensure that ambient air samples are collected in the prescribed manner and to

ensure that target compound qualitative and quantitative analyses are performed with known

precision and accuracy. Data quality objectives for the UATMP are presented in Table 4-3. The

data quality objectives for PAMS ambient air canister analyses are the same as those described

for the SNMOC and the UATMP (flagged with a @) and summarized in Tables 4-2 and 4-3,.

respectively.

The following canister pressure acceptance criteria have been adopted for the UATMP

program:

• . Based on previous experience,(4"10'12"16) no upper limits on the canister pressure are obtained from a UATMP sampling site. If the initial vacuum is less than 0.5 inches Hg, however, the sample is flagged on the data sheet, chromatogram, and log book.

0121.00 A7

1 March 2000

1 of 15

glp/D:\SECT4.WPD

Table 4-1

m o NMOC Data Quality Objectives "a o

QC Check Frequency Acceptance Criteria Corrective Action

Calibration Check (midpoint of curve)

Daily Relative percent difference within 20% of average calibration response (avg-daily)/avg

1) Repeat analysis of the point

2) Repeat analysis at different level

3) Repeat calibration curve 4) Remake and reanalyze

standard

System blank - Wet Air <50% %RH

Daily (after a calibration check)

<10.0ppbC 1) Repeat analysis 2) Leak check system 3) Notify task leader

Multi-point Calibration; 5 point plus zero, 3 injections per point

At the beginning and end of the sampling season

Correlation criteria (r2) > 0.995. Each point must have an RSD <3% (except zero)

1) Repeat one or two individual points

2) Repeat entire curve 3) Remake and reanalyze

curve

Replicates All duplicates Within 100 area counts - RSD ±5%

1) Notify coordination director

Can Cleaning One can analyzed on the Air Toxics system per batch of eight-highest total NMOC

Less than 10.0 ppbC 1) Repeat analysis once 2) Reclean canister batch

•TJ O ^ M ^

•2. CD o O 3 rt

O

o

O O o

o

> o o

Table 4-2

Summary of SNMOC Procedures

QC Check Frequency Acceptance Criteria Corrective Action

System Blank Analysis Daily, following calibration check

20 ppbC total 1) Repeat analysis 2) Check system for leaks 3) Clean system with wet air

Multiple point calibration (5 points minimum); propane bracketing the expected sample concentration.

Prior to analysis and monthly Correlation Coefficient (r)>0.995

1) Repeat individual sample analysis

2) Repeat linearity check 3) Prepare new calibration

standards and repeat

Calibration check using midpoint of calibration " curve spanning the carbon range (C,-C10)

Daily on the days of sample analysis

Response for selected hydrocarbons spanning the carbon range within ±30% difference of calibration curve slope

1) Repeat check 2) Repeat calibration curve

Replicate analysis All duplicate field samples Total NMOC within ±30% RSD

Repeat sample analysis

Canister cleaning certification

One can analyzed on the Air Toxics system per batch of eight-highest total NMOC

< 10 ppbC total Reclean canisters and reanalyze

Table 4-3

Air Toxics TO-15 QC Procedures

QC Check Frequency Acceptance Criteria Corrective Action

Bromofluorobenzene (BFB) Instrument Performance Check®

Daily* prior to calibration check and sample analysis

Analyst evaluates by HP Chemstation®; criteria can be found in software

1) Retune 2) Clean ion source and/or

quadrupoles

Five point calibration bracketing the expected sample concentration

Following any major change, repair or maintenance if daily QC is not acceptable. Recalibration not to exceed six weeks.

1) Relative Standard Deviation (RSD) of response factors <30%

2) Relative Retention Times (RRT) for target peaks ±0.06 units from mean

. relative retention time

1) Repeat individual sample analysis

2) Repeat linearity check 3) Prepare new calibration

standards and repeat analysis

Calibration check using mid-point of calibration . curve or one other point in curve®

Daily* on the days of sample analysis

Analyst verifies that the response factor <30% bias from calibration curve average response factor

1) Repeat calibration check 2) Repeat calibration curve

System Blank Analysis® Daily* following BFB and calibration check; prior to analysis

1) 0.2 ppbv per analyte or the MDL, whichever is greater

2) Internal Standard (IS) area response ±40% and IS Retention Time (RT) ±0.33 min. of most recent calibration check

1) Repeat analysis with new blank can

2) Check system for leaks, contamination

3) Reanalyze blank

Laboratory Control Standard (LCS)

Daily* 1) Recovery Limits 70% - 130% 2) IS RT ±0.33 min. of most recent

calibration

1) Repeat analysis 2) Repeat calibration curve

Replicate Analysis® All duplicate field samples

<30% Relative Percent Difference (RPD) for compounds greater than 5 times MDL

1) Repeat sample analysis

Table 4-3

(Continued)

QC Check Frequency Acceptance Criteria Corrective Action

Canister Cleaning Certification

One can analyzed on the Air Toxics system per batch of eight-highest total NMOC

<0.2 ppbv per VOC targeted compounds or MDL, whichever is greater

1) Reclean canisters and reanalyze

Sampler Certification Annual 1) Recovery 80% to 120% of targeted compounds for certification challenge

2) <0.2 ppbv or the MDL whichever is greater of targeted compounds for blank certifications

1) Repeat certification of canisters

Samples® All samples IS RT±0.33 min. of most recent calibration validation

1) Repeat analysis

*Every 24 hours frequency. ®QA criteria also needed for SNMOC and PAMS analysis.

Project No. Element No. Revision No. Date Page

0121.00 A7

1 March 2000

6 of 15

• If the elapsed time for sample collection is less than 20 hours or greater than 28 hours, the sample will be flagged on the data sheet, the chromatogram, and log book.

• Because the analytical system cannot extract a sample from a vacuum greater than 10 inches Hg, any canisters received with vacuum greater than 10 inches Hg will be ' voided.

• If the final canister vacuum is between ,7 and 10 inches Hg, the analysis is performed, but the sample is flagged on the data sheet, the chromatogram, and in the log book. If two successive samples from the same site result in canister vacuums greater than 11 inches Hg for each canister, the site operator is contacted and appropriate corrective action taken.

• For duplicate samples, if one or both canisters have a final vacuum between 7 and 10 inches Hg, only the canister with the higher vacuum will be analyzed, but the analysis cannot be replicated. If one of the duplicate canisters has a final vacuum higher than 15 inches Hg, neither is analyzed.

• If the duplicate samples have initial canister pressures that differ more than 0.5 inches Hg, only the canister with the higher vacuum will be analyzed and the occurrence recorded in the sample log book.

Quality objectives determined for the carbonyl analysis and the HAPs supported carbonyl

compounds listed in Category II analytes in Table 3-5 are to ensure that ambient air samples are

collected in the prescribed manner and to ensure that compound quantitative analyses are

performed with known accuracy and precision. The data quality objectives for carbonyl analysis

are presented in

Table 4-4.

Quality objectives determined for Semivolatile organic compounds (Category III) and

Polynuclear Aromatic Hydrocarbons (PAHs, Category IV) are-to ensure that ambient air samples

are collected in the prescribed manner and to ensure that target compound quantitative analyses

are performed with known precision and accuracy. The data quality objectives for these

compounds are presented in Table 4-5.

glp/D:\SECT4.WPD

Table 4-4

Carbonyl Data Quality Objectives

Parameter Quality Control

Check Frequency Acceptance Criteria Corrective Action

HPLC Column Efficiency

Analyze second source QC sample (SSQC)

At setup and 1 per sample batch

Resolution between acetone and propionaldehyde > 1.0 Column efficiency > 5,000 plate counts

Eliminate dead volume, back flush, or replace the column, repeat analysis

Linearity Check

Run a 5-point calibration curve and SSQC in triplicate

At setup or when calibration check is out of acceptance criteria

Correlation coefficient > 0.999, relative error for each level against calibration curve ± 20% or less Relative Error

Check integration, reintegrate or recalibrate

Linearity Check

Run a 5-point calibration curve and SSQC in triplicate

At setup or when calibration check is out of acceptance criteria

Intercept acceptance should be < 10,000 area counts per compound which correlates to 0.06 mg/mL

Check integration, reintegrate or recalibrate

Retention Time Analyze calibration midpoint

Once per 10 samples Acetaldehyde, Benzaldehyde, Hexanaldehyde within retention time window established by determining 3o or ±2% of the mean calibration and midpoint standards, whichever is greater

Check system for plug, regulate' column temperature, check gradient and solvents

Calibration Check

Analyze midpoint standard

Once per 10 samples 85-115% recovery Check integration, recalibrate or. remake standard, reanalyze samples not bracketed by acceptable standard

Calibration Accuracy

SSQC Once after calibration in triplicate

85-115% recovery Check integration, recalibrate or remake standard, reanalyze samples not bracketed by acceptable standard

Calibration Accuracy

Analyze O.lug/mL standard

Once after calibration in triplicate

±25% difference

Check integration, recalibrate or remake standard, reanalyze samples not bracketed by acceptable standard

System Blank Analyze acetonitrile Bracket sample batch, 1 at beginning and 1 at end of batch

Measured concentration < 5 times the MDL Locate contamination and document levels of contamination in file

Duplicate Analyses

Duplicate samples As collected ±20% difference Check integration, check instrument function, reanalyze duplicate samples

Replicate Analyses

Replicate injections Duplicate samples only <, 10% RPD for concentrations greater than 1.0 pg/mL.

Check integration, check instrument function, reanalyze duplicate samples

Table 4-4

(Continued)

Parameter Quality Control

Check Frequency Acceptance Criteria Corrective Action

Method Spike/Method Spike Duplicate (MS/MSD)

Analyze MS/MSD One MS/MSD per 20 samples

80-120% recovery for all compounds. Check calibration, check extraction procedures

•a era CD

CD <

?o w J o S3- is 2 £ .

CD O o

3

O

CD

O

2 o

a oo o »-+>

Ul o ^

to o o o

o to

> - J ;

o o

Table 4-5

Quality Control Procedures for Analysis of Semivolatile Organic Samples According to EPA Method 8270

Quality Control Check Frequency Acceptance Criteria Corrective Action

Decafluorotripheny 1-phosphine (DFTPP) instrument tune check

Daily prior to calibration check and sample analysis; every 12 hours if instrument is operated 24 hours/day

Evaluation criteria in Table 3 of Method 8270

1. Re-tune instrument; re-analyze 2. Clean ion source; re-tune

instrument; re-analyze 3. Prepare new tune check standard;

analyze

Five-point calibration Following any major change, repair, or maintenance if daily quality control check is not acceptable. Minimum frequency every six weeks, more frequently if required

Average Relative standard deviation (RSD) of response factors for all compounds should be <30%;. average RSD for Calibration Check Compounds must be <30%

1. Repeat individual sample analyses 2. Check calculations 3. Perform maintenance on GC,

especially leak check 4. Clean ion source 5. Prepare new calibration standards

and repeat analysis

System Performance Check Compounds (SPCCs) •

Daily (or every 12 hours) Minimum response for SPCCs of0.050

1. Repeat individual sample analyses 2. Check calculations 3. Perform maintenance on GC,

especially leak check 4. Clean ion source 5. Prepare new calibration standards

and repeat analysis

Calibration Check Compounds (CCCs)

Daily (or every 12 hours) Percent difference for each compound must be less than 30% relative to the mean of the calibration curve

1. Repeat individual sample analyses 2. Check calculations 3. Perform maintenance on GC,

especially leak check 4. Clean ion source 5. Prepare new calibration standards

and repeat analysis

Reagent Blank Once per 20 samples (5%) All analytes <5 x Method Detection Limit

1. Repeat analysis 2. Flag data

Table 4-5

(Continued)

Quality Control Check Frequency Acceptance Criteria Corrective Action

Surrogate compound Every sample 1. Repeat analysis recoveries 2. Flag data nitrobenzene-d5 35-114% 2-fluorobiphenyl 43-116% /?-terphenyl-d14 33-141% phenol-d6 10-94% 2-fluorophenol 21-100% 2,4,6-tribromo- 10-123%

phenol

h3

CD

fr) '"d

< O P ^

o tr K> O o o

o

W O ' - '

o

> - J

o o

Project No. Element No. Revision No. Date

0121.00 A7

1 March 2000

11 of 15 Page

The data quality objectives for phosgene, 2,3,7,8-tetrachlorodibenzo-p-dioxin, and

ethylene oxide (Category III) from the HAPs Table 3-5 are to ensure that ambient air samples are

collected in the prescribed manner and to ensure that target compound qualitative and

quantitative analyses, are performed with known precision and accuracy. The data quality

objectives for phosgene, 2,3,7,8-tetrachlorodibenzo-p-dioxin, and ethylene oxide are listed in

Tables 4-6, 4-7, and 4-8, respectively.

Quality objectives determined for the Clean Air Act metals (Inorganic HAPs, Category V

from Table 3-5) are to ensure that ambient air samples are collected in the prescribed manner and

to ensure that compound quantitative analyses are performed with known accuracy and precision.

The data quality objectives for the metals analysis are presented in Table 4-9.

glp/D:\SECT4.WPD

Project No. Element No. Revision No. Date Page

Table 4-6

Quality Control Parameters for Ethylene Oxide Analysis Performed According to the Analytical Procedures of NIOSH Method 1614

Parameter Frequency Acceptance Criteria Corrective Action

Six-point calibration Initially - prepare with samples

Checked by blind spikes and analyst spikes

1. Re-analyze individual calibration standards

4. Re-prepare calibration standards; re­analyze

Calibration check I -(Analyst spikes)

Daily - analyze 3 Relative percent difference between compound and mean response factor from calibration curve <20%

1. Re-analyze calibration check I

2. Check calculations 3. Re-prepare analyst

spike 4. Re-calibrate

Calibration check I I -(Blind spikes)

Daily - analyze 3 Relative percent difference between compound and mean response factor from calibration curve 5; 20%

1. Re-analyze calibration check I I

2. Check calculations 3. Re-calibrate

Reagent blank, Laboratory,blank

One per batch of samples

No analyte present above method detection limit

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

Measure breakthrough for each sample

All samples If the sample back analysis is > one tenth of the sample front analysis. Wb>Wf/10

1. Report breakthrough and possible sample loss.

0121.00 A7

1 March 2000

12 of 15

glp/D:\SECT4.WPD

Project No. 0121.00 Element No. A7 Revision No. 1 Date March 2000 Page 13 of 15

Table 4-7

Quality Control Parameters for Dioxin/Furan Analysis Performed According to the Analytical Procedures of EPA Method 8290

Parameter Frequency Acceptance Criteria Corrective Action

Five-point calibration Initially; repeat when • daily calibration check does not meet acceptance criteria

Relative standard deviation for response factors <20%, <30% for labeled reference compounds

1. Re-analyze individual calibration standards

2. Re-tune HRMS 3. Clean ion source 4. Re-prepare

calibration standards; re­analyze

Calibration check Daily (or every 12 hours if instrument is . operated 24 hrs/day)

Relative percent difference between compound and mean response factor from calibration curve <25%, <30% for labeled reference compounds

1. Re-analyze calibration check standard

2. Check calculations 3. Re-calibrate

Method spike/method spike duplicate

One per twenty samples (5%)

Accuracy: ± 30% Precision: ±50%

1. Repeat analysis 2. Flag data

Reagent blank, Laboratory blank

One per twenty samples (5%)

No analyte present above method detection limit

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

Field duplicate One per twenty samples (5%)

No analyte present above method

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

Trip blank One per twenty samples (5%)

No analyte present above method

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

glp/D:\SECT4.WPD

Project No. 0121.00 Element No. A7 Revision No. 1 Date March 2000 Page 14 of 15

Table 4-8

Quality Control Parameters for Phosgene Performed According to the Analytical Procedures of Compendium TO-6

Parameter Frequency Acceptance Criteria Corrective Action HPLC Column Efficiency

Analyze second source QC sample (SSQC)'

At setup and 1 per sample batch

Five-point calibration

Five-point calibration Initially; repeat when daily calibration :check does not meet acceptance criteria

Relative standard deviation for response factors <20%, <30% for labeled reference compounds

1. Re-analyze individual calibration standards

4. Re-prepare calibration standards; re­analyze

Calibration check -intermediate concentration standard near anticipated'levels -at least 10 times the detection limit

Daily Relative percent difference between compound and mean response factor from calibration curve zl0%

1. Re-analyze calibration check . standard

2. Check calculations 3. Re-prepare

calibration standard 4. Re-calibrate

Method spike/method spike duplicate

One per batch of samples (5%)

Accuracy: ± 30% Precision: ±50%

1. Repeat analysis 2. Flag data

Reagent blank, Laboratory blank

One per batch of samples (5%)

No analyte present above method detection limit

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

Replicate Analysis One per batch of samples

Relative standard deviation of ±15-20%

1. Repeat analysis 2. Flag data

Precision and Recovery Once per year Recovery and Precision comparable as listed: Cone, (ppbv) 0.034 63% 13 std 0.22 87% 14 std 3.0 99% 3 std 4.3 109% 12 std 20 99% 14 std 200 96% 7 std

1. Repeat analysis 2. Re-prepare reagent

blank; analyze 3. Flag data

glp/D:\SECT4.WPD

Table 4-9

Quality Control Measures for Metals Analysis According to Method 10-3.5

Parameter Frequency Acceptance Criteria Corrective Action

Multipoint calibration Daily Correlation coefficient > 0.995

1. Repeat analysis of calibration standards

2. Re-prepare calibration standards and re-analyze

Calibration check Daily Recovery 95-105% for all analytes

1. Repeat analysis of calibration check standard

2. Repeat analysis of calibration standards

3. Re-prepare calibration standards and re-analyze

Continuing calibration verification Every 10 samples Recovery 90-110% 1. Repeat analysis of continuing calibration verification sample

2. Reprepare continuing calibration -verification sample and re-analyze

3. Reanalyze samples since last acceptable continuing calibration verification

Method blanks Every 10 samples Analytes below method detection limit

1. Reanalyze 2. Reprepare blank and re-analyze 3. Correct contamination and reanalyze

blank 4. Repeat analyses of all samples since

last clean blank

Laboratory control sample One per sample batch Recovery 80-120% Reprepare sample batch; re-analyze

Method spike/method spike duplicate one per sample batch Recoveries 80-120% Flag data.

ha o £d M c£ £ 3 § £

" (1 o O 3 ~ 3 z z Z % ?

-1 O K>

O ' >— o

o

o o

Project No. Element No. Revision No. Date March 2000

1 of 2

0121.00 A8.

Page

SECTION 5

SPECIAL TRAINING REQUIREMENTS/CERTIFICATION

This program is an ambient monitoring program, performed using recognized EPA

sampling and analytical protocols and requiring the efforts of field sampling personnel and

analytical laboratory staff.

5.1 Sampling Personnel

Sampling personnel involved in this project have been trained in their tasks and have

from 1 to 25 years of experience in the duties they will be performing in the field. The field

testing staff will be subject to on-site surveillance by the EPA and ERG Task Leader with

appropriate corrective action enforced, if necessary. ERG personnel setting up the sampling

equipment will also be subject to on-site surveillance by the ERG Task Leader with appropriate

corrective action enforced, if necessary. ERG provides employee training, with specialized,

in-house training classes and on-the-job training by supervisors and co-workers. The monitoring

sites may be inside a sampling building or outside. There are no unusual hazards and no special

safety training or equipment required. All sampling staff will follow the ERG Health and Safety

Plan. The ERG Task Leader will pay special attention to potential heat or pollutant exposure on

a daily basis as conditions change at the site.

5.2 Analytical Laboratory Personnel

Analytical laboratory personnel involved in this project have been trained in their tasks

and have from 1 to 25 years of experience in the duties they will be performing in the analytical

laboratory. Laboratory staff will be subject to on-site surveillance by the Quality Assurance

staff. The samples involved in this program are being generated by monitoring of air emissions.

No unusual hazards are expected and no special safety training or equipment will be required to

perform the analyses. The laboratory will adhere to the ERG Health and Safety manual.

glp/D:\SECT5.WPD

Project No. Element No. Revision No. Date Page

0121.00 A8

1 March 2000

2 of 2

glp/D:\SECT5.WPD

Project No. Element No. Revision No. Date

0121.00 Bl

1 March 2000

1 of 8 Page

B—MEASUREMENT DATA ACQUISITION

SECTION 6

SAMPLING PROCESS DESIGN

Sampling procedures for the NMOC and UATMP programs are discussed in this section.

ERG provides site-specific support for the PAMS and HAPs sampling.

6.1 NMOC and SNMOC Sampling

Sampling takes place each workday from the beginning of June to the end of September

at NMOC and SNMOC sites from 6:00 a.m. to 9:00 a.m., standard time. Sampling procedures

have been discussed in detail in other documents.0'10 Figure 6-1 is a diagram of the sampling

system used for collecting the ambient air samples. Evacuated stainless steel canisters are

shipped daily from ERG's Research Triangle Park (RTP) Laboratory to the NMOC and SNMOC

sites. Canisters are connected by local operators to the sampling system as shown. The timer

will automatically activate the pumps and solenoid valve to begin and complete the sampling.

The Metal Bellows®-pump will pressurize air samples during the sampling period to about

15 psig, and the critical orifice will operate at sonic velocity to ensure a constant sampling rate

over the 3-hour period (a 21 micron stainless steel filter is installed in the sampling line and

removes particulate from the ambient air that may damage or plug the critical orifice). The

sample intake point ranges from 3 to 10 meters above ground level.

ERG installs the site sampling systems and trains designated local operators on-site. It is

the responsibility of the local operators to operate the sampling apparatus and complete the field

sample data form that ERG supplies with each canister. ERG staff maintains telephone contact

throughout the project to provide whatever assistance is needed to solve technical problems that

occur during the course of the program.

glp/D:\SECT6.WPD

Filter O Timer

• Sample inlet is an Inverted glass funnel. Lines and fittings are stainless steel.

Out

Solenoid Latching Valve

Metal Bellows® Pump MB151

Pressure Gauge

9

Canister(s)

Figure 6-1. NMOC, SNMOC, and 3-Hour Air Toxics Sampling System Components

Project No. Element No. Revision No. Date

0121:00 Bl-

1 March 2000

3 of 8 Page

ERG creates a sampling schedule, including the appropriate number of samples, when

sites are specified and site requirements are established.

All NMOC and SNMOC sites are usually scheduled to begin sampling at the beginning

of June and continue to the end of September. With a 3-hour ambient air sample, both PDFID,

SNMOC, and air toxics measurements may be performed on the same canister i f enough pressure

remains in the canister. It is recommended that any aliquots for analysis be taken from the

canister on successive days to allow equilibration between analyses.

6.1.1 Air Toxic Compounds Sampling

The 3-hour air toxics samples under the NMOC program are analyzed from the same

canisters as the NMOC and/or SNMOC samples. Refer to Section 6-2 for sampler certification.

6.1.2 Carbonyl Compounds Sampling

Carbonyl samples are collected using DNPH cartridges with an integrated sampling

system (e.g., stand-alone pumps, capillary critical orifices, ozone scrubbers ahead of the

DNPH cartridges), shown in Figure 6-2.

6.2 UATMP Sampling

Prior to installation of the UATMP sampler at a site, the sampler is tested at the ERG

RTP laboratory for performance capability and qualified for cleanliness. Cleaned, humidified air

is flushed through the sampler for at least 48 hours to remove organic contaminants in the

system. The cleaned, humidified air is then analyzed and the results placed in a permanent file to

record any contamination following EPA Compendium Methods TO-14A and TO-15. The

glp/D:\SECT6.WPD

Project No. Element No. Revision No. Date Page

0121.00 Bl

1 March 2000

4 of 8

Power Line

Sample Inlet

Thermocouple Line

II Duplicate 11 f j Sample .' _Q

Cartridge '

Calibrated Rotometer Duplicate ( V )

Orifice V

Isolation ' \ ~ 7 >

Valve

Sample Cartridge

0 Orifice

Elapsed Timer

In

Pump g

glp/D:\SECT6.WPD

Figure 6-2. Carbonyl Sampling System

Project No. 0121.00 Element No. Bl Revision No. 1 Date March 2000

. Page 5 of 8

samplers are then challenged with a mixture of known concentration to qualify the sampler.

These results are placed in a permanent file.

ERG will establish a schedule for the UATMP sites when the sites are identified. A total

of 30 sampling days will be scheduled per site program and will be identified in the schedule.

Days for duplicate sampling will be designated.

Integrated ambient air samples are taken in 6-liter stainless steel SUMMA®-treated

canisters for a 24-hour period beginning at midnight. Cleaned canisters are shipped to the site

under vacuum from the ERG RTP laboratory. After sampling, the final desired pressure in the

canister should be between 2 and 5 inches Hg vacuum.

The sampling assembly for the UATMP is shown in Figure 6-3. The driving force for

filling the canister is its initial vacuum. The single-head purge pump shown in Figure 6-3 is used

to purge the sample inlet lines and to draw ambient air through the carbonyl sampling probe and

cartridges.

6.2.1 Carbonyl Compounds Sampling

Carbonyl sampling occurs at UATMP sites at the same time the canister samples are

taken. DNPH sampling cartridges are connected to the UATMP sampler as shown in

Figure 6-3 when the 6-liter canisters are connected, and ambient air is drawn through the

cartridges through a separate heated sampling probe. Each DNPH cartridge has an ozone

scrubber (Figure 6-4) in the sample manifold to remove ozone before the ambient air sample

enters the DNPH cartridge. The ozone scrubbers are replaced before each season (yearly).

Purchased DNPH cartridges are shipped to each site for carbonyl sampling. A total of 34 tubes

will be analyzed per site, including ten percent duplicate samples and ten percent field blanks per

season.

glp/D:\SECT6.WPD

cro CD

o ^ w 5 g -2. CD

O 13

Z O

3 ~ z Z o o •

2 - micron Particulate Fitter

o/s/g/mor/33961099rt4-2.tif

O

CO

o

o o o

o

W o

Figure 6-3. Sampling Assembly for the UATMP

Project No. Element No. Revision No. Date Page

0121.00 Bl

1 March 2000

7 of 8

Sectional View

~ 3" Potassium Iodide Coated 1/4" O.D. Copper Tubing

0^s/g/morrls/379?/p9ms/hobson/Hob1.eps

o:/s/g/morris/3797/pams/hobson/hobl.ppt

Figure 6-4. Cross-Sectional View of the Ozone Scrubber Assembly

glp/D:\SECT6.WPD

/ Project No. . 0121.00 Element No. Bl Revision No. 1 Date March 2000 Page 8 of 8

ERG ships DNPH sampling cartridges to each site in the shipping container with the

6-liter canister(s). The carbonyl samples are also collected for a 24-hour period. After sampling,

the cartridges are removed from the sampling apparatus, sealed, and returned to the ERG RTP

laboratory in the shipping container with the canister(s). Disposable polyethylene gloves are

used by the field operators when handling the cartridges to reduce background contamination

levels. Additional details of the carbonyl sampling and analysis procedures are presented in the

EPA Compendium Method TO-11 A.

6.3 PAMS Sampling

PAMS sampling is performed completely by the PAMS sites in accordance with the

TAD, ( 3 ) with ERG supplying only such support as requested (e.g., sampling system and training,

automated GC systems). ERG ships cleaned canisters and prepared carbonyl compounds

sampling cartridges to the PAMS sites on the appropriate schedule to support the sampling

program, and the samples are shipped to the ERG RTP laboratory for analysis. Exact provision

for support of automated GC systems is site specific; each site may work with Chromlan during

the sampling season.

6.4 HAPs Sampling

HAPs sampling is performed completely by the sites in accordance with the methods

listed in Table 3-5. ERG receives the samples from the sites for analysis only.

glp/D:\SECT6.WPD

Project No. Element No. Revision No. Date

1 March 2000

1 of 17

0121.00 B2

Page

SECTION 7

SAMPLE HANDLING AND CUSTODY REQUIREMENTS

Similar canister sample custody procedures are followed for all monitoring programs;

however, program-specific differences exist because the canister cleanliness requirements and

the analytical requirements for the three programs vary.

7.1 NMOC, SNMOC, and UATMP Sample Custody

7.1.1 NMOC Sampling Field Data Forms

A color-coded, three-copy canister sample data sheet (Figure 7-1) is shipped with each

6-liter canister to an NMOC/SNMOC or UATMP site. If duplicate samples are to be taken, two

canisters and two data sheets are sent in the shipping container to the site. When a sample is

taken, the site operator fills out the field data form according to the instructions in the

NMOC/SNMOC or UATMP on-site notebook. The site operator detaches the pink copy, inserts

it in the on-site notebook, and sends the remaining copies with the canister in the shipping

container to ERG's laboratory.

Upon receipt, the sample canister vacuum/pressure is compared against the field

documented vacuum/pressure to ensure the canister remained airtight during transport. If any

leaks are detected, the sample is voided. The canister information is then entered electronically

into the computer login (login information is shown in Figure 7-2), given a unique ERG

identification (ID) number, and tagged (see Figure 7-3), noting the site location and the sample

collection date. The samples are also logged into the computerized login database. The

remaining copies of the canister sample data sheet are separated; the white copy is stored with

the

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000

2 of 17

lERG t i "

Canister Sample Data Sheet LAB ID # EASTERN RESEARCH GROUP, INC;

Q _ l uu

o z > j j o 111

Location: City / State _ Sampling Period: Nox Analyzer Operating: Average PPM:

Elapsed Time:

Average Wind Speed: _ Average Wind Direction: Average Temperature: _ Average Barometric Pressure: Relative Humidity: Flow Controller Set at: Comments:

Site Code: Collection Date:

Canister Number: Operator:

Initial Vacuum: Final Field PressureA/acuum:

Duplicate (Y/N) Duplicate Can #:

Options: Flow Controller Zero Reading:

Received by: •_ Date Received: Carbonyl Tubes:

Carbonyl ID #:

Pressure @ Receipt: Void Acceptable Yes No

Stored:

o o

Analyst: Analysis Date: Analysis Time: NMOC Instrument:

Area Counts run 1: ppmC run 1:

Canister Number: Analysis Pressure: Sample Replicate:

Initial or Repeat:

Area Counts run 2: ppmC run 2:

Average AC: Standard Dev:

Entered into Database by:

Average ppmC: Standard Dev:

Area Counts run 3: ppmC run 3:

Date:

o O s z

Analyst: Analysis Pressure: Load Volume:

Date: Data File Name: Duplicate File Name:

Date:

V) o X o

Analyst: Analysis Pressure: Load Volume:

Date: Data File Name: Duplicate File Name: Replicate File Name:

White: Sample File Copy Yellow: Receiving Copy Pink: Field Copy

Figure 7-1. Canister Sample Data Sheet

glp/D:\SECT7.WPD

Sample Login

ERG Sample ID Date

Received Rec'd

By Date

Sampled Canister Number

coc Present

Tubes Present Project Name & Sample Description

Storage Location

ERG Sample ID Date

Received Rec'd

By Date

Sampled Canister Number

coc Present

Tubes Present Project Name & Sample Description Cans Tubes

* •

-

Figure 7-2. Sample Receipt Login Information *n o w I_d

™. CD O O 3 rt

^ 5 2

3 o tr

O to O

>— o - J o

o to

o o

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000

4 of 17

o

Site:

Date:

I.D. #

Channel PPMC

Figure 7-3. Canister Tag

canister until analysis is complete and the yellow copy is stored chronologically in the receiving

notebook. The sample ID number is written on the canister tag and on all ERG copies of the data

sheet.

7.1.2 NMOC Invalid Sample Forms

The canister sample data sheet may indicate that the sample sent from a site is invalid.

When a sample is designated as invalid, the assigned ERG ID number is voided and an NMOC

invalid sample form (Figure 7-4) is attached to the data sheet. The sites will be notified in the

analytical reports of any invalid samples. I f the site seems to have problems taking a valid

sample, normally two voids in a row, the site task leader will work with the site personnel to

eliminate the problem.

glp/D:\SECT7.WPD

Project No. Element No. Revision No Date Page

*ERG EASTERN RESEARCH GROUP. INC.

INVALID SAMPLE FORM Site Code: .__

City: ; State:

Sample Collection Date: ; Operator:

Sample Canister Number: •

Sample Duplicate for this Date: Yes • No •

If Yes, Duplicate Canister Number: .

Reason for Invalid or Missed Sample:

0121.00 B2

1 March 2000

5 of 17

Average NOx Analyzer Reading for this Collection Date:

Wind Speed: Wind Direction: •

Rotameter Numbers: . 'Rotameter Indicated Flow Rate:

Average Barometric Pressure (mm Hg or inches Hg):

Ambient Temperature (°F): Relative Humidity:

Sky/Weather Conditions:

Received By: Date: ._ Action Taken:

Resolution:.

Field Invalid or In-house Invalid

Figure 7-4. NMOC Invalid Sample Form

glp/D:\SECT7.WPD

Project No. 0121.00 Element No. B2 Revision No. 1 Date March 2000 Page 6 of 17

7.1.3 NMOC Sample Analysis Fonns

The ERG NMOC analyst completes the canister sample data sheet-NMOC section and

NMOC daily HP5880 calibration form (Figure 7-5), which must include the following items:

• Critical instrument parameters (check-list format)

• Sample canister number

• Analysis date

Sequential ERG ID

• The analyst's name

• Calibration cylinder used

• Analysis start time

• Results of the NMOC analysis (individual replicates and NMOC average)

The information from the daily calibration form is added to the computer data file.

NMOC daily HP5880 Calibration forms are filed consecutively by ERG Sample ID number in a

three-ring analysis notebook for permanent record.

7.1.4 NMOC Canister Log

All canisters are cleaned prior to reuse using ERG SOP-MOR-062. All canisters,

whether used for NMOC, UATMP, or PAMS, are cleaned by the same procedure and are entered

into the canister cleanup log, shown in Figure 7-6.

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000

7 of 17

*EKG E A S T E R N R E S E A R C H G R O U P . I N C .

Date:

FID Instrument (A-D):

'Hydrogen Pressure:

Air Pressure:

NMOC Daily HP5880 Calibration Form

. Analyst

Propane Calibration Cylinder No.:

Label Concentration:

Actual Concentration:

ppmC Propane

Initial Daily Calibration:

Time:(DST) ;

Zero Air AC

X =

Rropane AC

Calibration Factor ppmC Propane

(Propane AC - Zero Air AC)

ppmv

ppmv

Final Calibration:

Time:(DST) _

Zero Air AC

[(

Propane AC

) - ( )]

X = > X =

Calibration Factor = ppmC Propane fProDane AC - Zero Air A O

j : , : } ; [( ) - ( )]

Figure 7-5. NMOC Daily HP 5880 Calibration Form

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date Page

Canister Cleanup Log

Can #

Cyc. 1 Cyc. 2 Cyc. 3 Pre

ppmC Post AC

Post ppmC

Final V

Date/ Initials Can # V P V P V P

Pre ppmC

Post AC

Post ppmC

Final V

Date/ Initials

V

!

« >

•

1

AC = Area Counts Figure 7-6. Canister Cleanup Log

glp/D:\SECT7.WPD

0121.00 B2

1 March 2000

8 of 17

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000

9 of 17

7.1.5 Canister Analytical Routing Schedule

The canisters received from the monitoring sites are placed in the laboratory by ERG

staff. The canisters are analyzed daily in the ERG RTP laboratory.

For the sites at which 3-hour air toxics will be analyzed, ten 3-hour air toxics samples per

site are selected from the NMOC samples at random. After NMOC analyses (PDFID), the

samples are sent to the ERG Air Toxics Laboratory for UATMP and speciated NMOC/PAMS

(GC/FID/MSD) analysis.

7.1.6 SNMOC/UATMP/PAMS Analysis Log

The SNMOC/UATMP/PAMS analysis log is shown in Figure 7-7 (Parts 1 and 2). The'

log is generic and is bound into a book with hard covers. The column headings on the log sheet

are given below, followed by a description of the information contained in the various cells for

the SNMOC, UATMP, or PAMS analyses:

UATMP sample identification.

UATMP site code.

DATA FILE NAME The data file number used by the Perkin Elmer

Turbochrom®, Analytical Software and the Hewlett Packard Chemstation® Software programs.

SAMPLE DATE Date the sample was taken.

ANALYSIS DATE Date the analysis was performed.

SAMPLE ID

FIELD ID

glp/D:\SECT7.WPD

</> m o UATMP Analysis Log

o Sample ID Field ID

Data File Name

Sample Date

Analysis Date

Standard Ref. No. Method

EM Voltage

. _.. - - -,- ,. - •- •• • -•• — • - - --

ere CD J2. CD o

o 3

o

3

2; o o •

o

C3 o cr to o o o

o to

to o o

Figure 7-7. UATMP Analysis Log

UATMP Analysis Log

Load Volume Can No. Analyst Comments Inches Hg Liters Can No. Analyst Comments

. _

Figure 7-7. (Continued)

STANDARD REF. NO.

METHOD

EM VOLTAGE

LOAD VOLUME

CAN NO.

ANALYST

COMMENTS

Project No. 0121.00 Element No. B2 Revision No. 1 Date March 2000 Page 12 of 17

Standard reference number. For samples and system blanks the column is left blank or indicated by "NA."

Method used to acquire data in HP Chemstation.

The electron multiplier voltage of the instrument.

Inches Hg - Canister pressure in inches of mercury. This column is not used because the volume is given in liters.

Liters - The load volume is recorded in liters according to the autosampling system.

Canister reference number

Analyst initials.

Any appropriate comments relative to the analysis.

7.2 Carbonyl Sample Custody

Figure 7-8 shows the multipage field data and custody sheet used for all carbonyl

sampling documentation. A field data sheet is shipped to the site with blank carbonyl tubes if the

tubes are provided by ERG, or blank data sheets are provided to sites supplying their own tubes

for sampling. After sampling, the field data sheet is completed by the site operator and a copy

retained for site records. The carbonyl sample tubes and field data sheet are shipped to the

analytical laboratory.

When samples are received, they are recorded in the sample receipt logbook (see

Figure 7-2), given an ERG sample ID number, and logged into the computerized database. The

database records each carbonyl sample and the field data sheet are put into a bag labeled with the

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000

13 of 17

*£RS E A S T E R N R E S E A R C H G R O U P . I N C .

Shipped

Recv'd

CARBONYL DATA SHEET

City Site Operator

AIRS No.

PAMS

SAMPLE DATE/TIME

SAMPLE DURATION

SAMPLE VOLUME LOT NUMBER

SAMPLE ID NUMBER

NMOC/TOXICS

Date

Lab ID

Duration

Rotameter Reading: @ set up

Rotameter Number

Lot Number

@ recovery

Volume (calculated by Lab)

Comments:

V

White - Lab Canary -' Receiving Pink - Sampler/Local Program

Figure 7-8. Carbonyl Field Data Sheet

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date

0121.00 B2

1 March 2000

14 of 17 Page

ERG ID number, site code, sampling date, individual tube designations, and date of receipt and

initials of receiving personnel. This sample bag is stored in a refrigerator designated for carbonyl

samples.

7.3 HAPs Sample Custody

Documentation while in the field monitoring phase of the program will use preformatted

forms. Field testing personnel will record data on a "Field Data Sheet." The field data sheet

provides for documentation of time, date, location, meteorological parameters and possibly some

laboratory parameters. Other documentation that will be used in the field is the identification

label shown in Figure 7-9. If Corrective Action is required during the portion of field

monitoring activities, the reason for the correction and action taken will be documented on the

"Corrective Action Report" (Figure 7-10). All forms will be written in indelible ink. I f

correction is required on the form, a single line will be drawn through the erroneous entry and the

correction will be dated and initialed. Any blank spaces will have a line drawn through to ensure

that the space is not filled in later. AH corrections will be authorized by the Site Coordination

Task Leader.

All analytical laboratories will provide sample tracking forms, narratives describing any

anomalies and any modifications to analytical procedures, data and sampling handling records,

and laboratory notes for inclusion in the final report. All laboratory electronic records will be

recorded for archive on magnetic media, and all hardcopies of raw data will be included in the

project archive file. *

glp/D:\SECT7.WPD

Project No. 0121.00 Element No. B2 Revision No. 1 Date March 2000 Page 15 of 17

Site Name

Site Address

Sampler Identification

Sample Type

Sampling Period

Date Collected

Signature

Figure 7-9. Label for Sample Identification

All records generated by measurement activities are signed or initiated by the person

performing the work and reviewed by an appropriate supervisor. Measurement results become

part of a project report which is reviewed by a technical reviewer. All notebooks are kept in

black ink, dated and signed by the person making the entries, and routinely inspected by the

appropriate supervision, as evidenced by his/her initials and data of inspection. Laboratory

notebook maintenance procedures are regulated by Standard Operating Procedures.

glp/D:\SECT7.WPD

Project No. 0121.00 Element No. B2 Revision No. 1 Date March 2000 Page 16 of 17

Corrective Action Report

Originator: Date:

Project Number: Corrective Action Number:

Description of Problem: (Give Date and Time Identified)

State Cause of Problem:

State Corrective Action Planned: (Include Persons Involved in Action)

QA Officer Comments:

Signatures: Project Manager Comments:

QA Officer:

Project Manager Comments:

Project Manager:

Project Manager Comments:

Originator:

Project Manager Comments:

Figure 7-10. Corrective Action Report

glp/D:\SECT7.WPD

Project No. Element No. Revision No. Date

0121.00 B2

1 March 2000

17 of 17 Page

7.4 Sampling Monitoring Data

All data sheets from the monitoring sites will be collected at the end of each monitoring

episode by the Task Leader and maintained in his custody throughout the monitoring program.

The data sheets will be released to the report writer after a thorough debriefing. The original

field data will remain in ERG custody and eventually stored on file with the final report.

glp/D:\SECT7.WPD

Project No. Element No. Revision No. bate

0121.00 B3

1 March 2000

1 of 17 Page

SECTION 8

ANALYTICAL METHODS REQUIREMENTS

Analytical procedures are program-specific because the instrumentation and the target

compounds of the four programs differ. The primary analytical instrument is PDFID for NMOC;

GC/FID/MSD for SNMOC, UATMP, VOCs and PAMS hydrocarbons; HPLC for carbonyls for

UATMP and PAMS; GC/MS for Semi volatile? and Phosgene; GC/ECD for Ethylene Oxide;

.HRMS for Dioxin/Furan; and ICPMS for Metals. All samples taken for SNMOC, UATMP, or

PAMS hydrocarbons can be evaluated for any of these parameters because the instrumentation is

currently collecting the same data.

8.1 Canister Cleanup System

A canister cleanup system (Figure 8-1) has been developed and is used to prepare sample

canisters for use and reuse after analysis (Standard Operating Procedure, ERG-MOR-062). An

oil-free compressor with an 80-gallon reservoir provides source air for the system. The

compressor was chosen to minimize hydrocarbon contamination. A coalescing filter removes

water mist and particulate matter down to a particle size of 10 microns and permeation dryers

remove water vapor from the compressor source air. The permeation dryers are used with a

moisture indicator to show detectable moisture in the air leaving the dryer.

Next, air is passed through catalytic oxidizers to destroy residual hydrocarbons. The'

oxidizers are followed by in-line filters for secondary particulate matter removal, and by

cryogenic traps to condense any residual water and organic compounds not destroyed by the

catalytic oxidizers. A single-stage regulator controls the final air pressure in the canisters and a

metering valve is used to control the flow rate at which the canisters are filled during a cleanup

cycle. The flow direction is controlled by a separate rotameter, installed in the clean, dried air

glp/D:\SECT8.-WPD

5.ty Filter Assembly

Air Compressor

Air Flow Rotameters

Dry Rotameter

_ . Air Cryolrap Pressure _ . . . Regulator Purge Valves

LU—1> ,_, Flow

Controller

2£

Catalytic Oxidizers

Cryolrap Purge Valve

Air Bypass -

Roughing

Pump

8-Port Manifold

D O O D D D D D D D O D D O D D

8Port Manifold

To Cert if leal ion System

o/&'g/mor/33961099T7-1 .b'f

A. Manifold Air Pressure Valve B. Manifold Vacuum Valve

C. Manifold Pressure Release Valve

D. Manifold Port for Connecting Canisters to be Cleaned

Figure 8-1. Canister Cleaning Apparatus

Project No. Element No. Revision No. Date

0121.00 B3

1 March 2000

3 of 17 Page

line. A shutoff valve exists between the clean dried air line and the humidifier system (which is

made up of a SUMMA® treated 6-liter canister partially filled with HPLC-grade water). One

flowmeter and flow-control valve route the cleaned, dried air into the 6-liter canisters, where it. is

bubbled through the HPLC-grade water; a second flow-control valve and flowmeter allow air to

bypass the canister/bubbler. By setting the flow-control valves separately, the downstream

relative humidity can be regulated. A setting of 100% relative humidity is used for canister

cleaning with the wet rotameter on and the dry rotameter off. Another shutoff valve is located

between the humidifier and each 8-port manifold where the canisters are connected for cleanup.

The vacuum system consists of a Precision Model DD-310 turbomolecular vacuum

pump, a cryogenic trap, an absolute pressure gauge, and a manifold vacuum valve connected as

shown in Figure 8-1. The cryogenic trap prevents the sample canisters from being contaminated

by back-diffusion of hydrocarbons from the vacuum pump into the cleanup system. The

manifold vacuum valves enable isolation of the vacuum pump from the system without shutting

off the vacuum pump. . . .

After sample analyses are completed, a bank of eight canisters is connected to each

manifold as shown in Figure 8-1, with each canister valve open and the air pressure, vacuum,

and bellows valves closed. The vacuum pump is started and one of the bellows valves is opened,

drawing a vacuum on the canisters connected to the corresponding manifold. After reaching

10 mm Hg absolute pressure, as indicated by the absolute pressure gauge, the vacuum is

maintained for 30 minutes. The bellows valves are then closed and the cleaned, dried air that has

been humidified is introduced into the evacuated canisters at a rate of 7.0 liters per minute until

the pressures reach approximately 20 psig. This flow rate has been recommended by the _

manufacturer as the highest flow rate at which the catalytic oxidizers can handle elimination of

.hydrocarbons with a minimum of 99.7% efficiency. The evacuation and pressurization of the

canisters constitutes one cleanup cycle.

glp/D:\SECT8.WPD

Project No. 0121.00 Element No. B3 Revision No. 1 Date March 2000 Page 4 of 17

The cleanup cycle is repeated twice more during the canister cleanup procedure.

Following the third pressurization, the canister valves are closed and the canister that had the

highest pre-cleanup concentration is selected for cleanliness verification. The cleanliness of the

canister is qualified by GC/MSD analysis (one canister per bank of cleaned canisters - one

canister per eight cleaned). Upon meeting the cleanliness criterion of the programs, the canister

is reconnected to the cleanup manifold. All canister valves are opened and the canisters are

evacuated to approximately 0.5 mm Hg absolute pressure a fourth time,-in preparation for

shipment to the site.

8.2 NMOC Analytical Systems

Two modified Hewlett-Packard Model 5880 dual-channel FID chromatographs are used

to determine the NMOC concentrations in the ambient air samples shipped daily to the ERG RTP

laboratory. Figure 8-2 shows a diagram of one NMOC system channel; four analytical channels

are designated as ERG Channels A, B, C, and D. A specific volume of sample is drawn from the

canister into a cryogenic trap and cooled with liquid argon. The NMOC fraction is condensed in

the sample trap. The 6-port valve is changed to the "Inject" position, the liquid argon is

removed, and the oven door of the chromatograph is closed. The oven is heated to 90°C at

30°C/min, and the NMOC is carried into the FID by the helium carrier gas. The results are

reported on the integrating recorder of the instrument. The analytical procedure described is the

PDFID method and is described in detail elsewhere.''•2'410)

Sites requesting 3-hour toxic analysis will have selected samples analyzed by

GC/FID/MSD. The analytical procedures for the GC/FID/MSD are described in Section 8.4. .

Sites requesting carbonyl analysis will have samples analyzed by the HPLC. The

analytical procedures for the HPLC are described in Section 8.5.

glp/D:\SECT8.WPD

Project No. Element No. Revision No. Date Page

0121.00 B3

1 March 2000

5 of 17

Absolute Pressure Gauge

Pressure Regulator

Vacuum Valve

Vacuum Pump

Dewar Flask

Glass Beads

Vent (excess)

1

Canister \ / Valve Y (optional fine

A ™ Canister

-O

1 1

Rotameter i FID

i

Integrator Recorder

Gas Purifier

Pressure Regulator

o/s/g/mor/33961099Zf7-2.tif

Figure 8-2. Schematic of Analytical Systems for NMOC glp/D:\SECT8.WPD

Project No. Element No. Revision No. Date

0121.00 B3

1 March 2000

6 of 17 Page

8.3 SNMOC Analytical Systems

The SNMOC and 3-hour toxic analysis'samples are analyzed by the same procedures

described for UATMP GC/FID/MSD in Section 8.4. The list of SNMOC target compounds is

shown in Table 3-1.

Sites requesting carbonyl analysis will have samples analyzed by the HPLC. The

analytical procedures for the HPLC are described in Section 8.5.

8.4 UATMP and Concurrent Analytical System

The UATMP GC/FID/MSD analyses are performed on a 400 mL sample from the

canister with a Hewlett Packard 5890 Series II GC and a Hewlett Packard 5971 Mass Selective

Detector using a 60 m by 0.32 mm ID and a 1pm film thickness J&W DB-1 capillary column

followed by a 2:1 splitter that sends the larger portion of the column effluent to the MSD and the

smaller fraction to the FID. Table 8-1 shows the GC/FID/MSD operating conditions. Figure 8-3

shows the GC/FID/MSD system arrangement. The list of UATMP target compounds is shown

in

The chromatograph oven, which contains the DB-1 capillary column, is cooled to -50°C

with liquid nitrogen at the beginning of the sample injection. This temperature is held for five

minutes and then increased at the rate of 15°C per minute to 0°C. The temperature is then

ramped at 6°C per minute to 150°C, then ramped at 20°C per minute to 225°C and held for

8 minutes. The gas eluting from the DB™-1 capillary column goes through the 2:1 fixed splitter,

which divides the flow to the MSD and FID.

The analytical procedure for UATMP carbonyl analysis performed by HPLC is described

in Section 8.5 for PAMS carbonyl analysis, '

Table 3-2.

glp/D:\SECT8.WPD

Project No. Element No. Revision No. Date Page

0121.00 B3

1 March 2000

7 of 17

Table 8-1

UATMP GC/FID/MSD Operating Conditions

Parameter Operating Value

Sample Volume 400 mL

J&W DB-l Capillary Column: Length Inside diameter Film thickness Oven temperature

60 m 0.32 mm

1 (im

-50°C for five minutes, 15°C/min to 0°C

then 6°C/min to 150°C, then 20°C/min to 225°C for 8

minutes

Temperatures: FID Injector Oven Temperature Auxiliary Temperature

300°C 210°C 278°C

Gas Flow Rates: Helium Carrier Gas Helium Make-up H 2 to FID Air to FID

1.5 mL/min 30 mL/min 30 mL/min

300 mL/min

Entech Sample Interface Conditions

Module 1 - Glass Bead Trap Initial Temperature Module 2 - Tenax® Trap Initial Temperature Module 3 - Cryofocuser Temperature

-180°C -40°C

-180°C

glp/D:\SECT8.WPD

6L S a m p l e Canisters

En tech 7000

Preconcen t ra to i

16-port Entech Autosampler

HP 5971 Mass Spect rometer

Detector

0.53 m m Silica C o a t e d

Guard C o l u m n

HP 5890 Series II Gas C h r o m a i o g a p h

Persona) C o m p u t e r suppl ied with

PE Turbocarbon for FID Peak Ident i f icat ion a n d Entech Software

Personal C o m p u t e r Suppl ied with

HP Chemsto t lon - for Mass Spectra

Ident i f icat ion

Nell son Analyt ical

A/D Inter face

Co lumn : J&WDB- l * Capi l lary C o l u m n

1 u FPm Thickness 6 0 m x 0 . 3 2 m m

Low D e a d Volume Stainless Steel Union

3:1 sp l l t -0 .5mi to FID-1 ml to MS

CSQ CD

a as CD ^

O 2

z o

tji ^} 1-1

3 a>' rt Z-o •

o o OO £3* i—'

O K ) ^

— o w o - J O i—' U ) o Figure 8-3. Gas Chromatograph/Mass Spectrometer/FID System

Project No. Element No. Revision No Date Page

8.5 PAMS Analytical Systems

The following analyses are included in the PAMS program.

The PAMS hydrocarbon samples are analyzed by the same procedures described for the

Concurrent GC/FID/MSD in.Section 8.4, with the target list shown in Table 3-3.

The PAMS and UATMP carbonyl samples are stored in the refrigerator after they are

received from the field prior to analysis. Sample preparation is performed by removing the

DNPH cartridge from its shipping vial and attaching it to the end of a 5-mL Micro-Mate glass

syringe. Four mL of acetonitrile is added to the syringe and allowed to drain through the

cartridge into a graduated centrifuge tube. After drainage has stopped, the extract is diluted with

acetonitrile to the 4-mL mark and mixed. This solution is then transferred to a 4-mL sample vial

fitted with a Teflon®-lined, self-sealing septum'and stored in a refrigerator until analysis is

performed.

The analytical separation of carbonyls is performed using either a Varian® 9000 HPLC or

a Waters HPLC configured with a 25 cm by 4.6 mm C-18 silica analytical column with a

5 micron particle size. A typical HPLC system is shown in Figure 8-4. ERG's system uses a •

Nelson Analytical A/D interface as the data system. Typically, 20-uL samples are injected with

an automatic sample injector. A mobile phase gradient of water, acetonitrile, and methanol is

used to perform the analytical separation at a flow rate of 1.0 mL/min. A multiwavelength

UV detector is .used at 360 nm.

0121.00 B3

1 March 2000

9 of 17

glp/D:\SECT8.WPD

m o

a

Column

Mobile Phase

OQ D ^ W J

2 o

3 CD

o

o

o

Figure 8-4. HPLC System o

- J

l - t o cr O o o

o

o o

Project No. Element No. Revision No. Date

0121.00 B3

1 March 2000

11 of 17 Page

8.6 Semivolatile Analytical Systems

Sampling modules containing XAD-2® and petri dishes containing filters, together with

Chain of Custody forms and all associated documentation, will be shipped to the ERG laboratory

from the field. Upon receipt at the laboratory, samples will be logged into the laboratory sample

tracking system and sent to the sample preparation laboratory. Sample preparation and analysis

procedures are based on SW-846 Method 3542 and SW-846 Method 8270.

To prepare the samples for analysis, a large Soxhlet extractor, containing approximately

500 mL of methylene chloride and several boiling chips is assembled. The sorbent from the

PS-1 sampling module is spiked with Method 8270 surrogate Compounds, and the XAD-2® is

placed in the Soxhlet extractor. Surrogate compounds are spiked into all of the samples,

including field sample and blanks. The filter from the associated petri dish is added to the top of

the XAD-2®. (If a field blank is to be prepared, clean XAD-2® and a clean filter from the same

batch as the filter and resin used for the field samples are used instead of the sampled resin and

filter.) The samples and blanks are extracted for 18 hours, and the round bottomed flask

containing the extracted sample is removed from the extractor.

A Kuderna-Danish apparatus with a 10 mL concentrator tube is assembled. A piece of •

pre-extracted glass wool is placed in a glass funnel and 20g of anhydrous Na2S04 is placed in the

inlet of the Kuderna-Danish apparatus. The round bottomed flask is rinsed three times into the

funnel with 10 mL aliquots of methylene chloride. A 3-bali macro Snyder column is pre-wetted

with methylene chloride and attached to the Kuderna-Danish inlet. The entire apparatus is placed

on a water bath at 85°C. The concentrator tube is half-immersed in the water bath, and the flask

is bathed in water vapor. The apparatus is removed from the water bath when an apparent'

volume of 5 mL is observed in the concentrator tube. The solvent is allowed to drain for

5 minutes. The Snyder column is rinsed with 1 mL of methylene chloride and allowed to drain

into the concentrator tube. The concentrator tube is removed from the evaporator, and a 2-ball

macro Snyder column is attached to the concentrator tube and pre-wetted with 2 mL of

glp/D:\SECT8.WPD

Project No. 0121.00 Element No. B3 Revision No. 1 Date March 2000 Page 12 of 17

methylene chloride. The apparatus is placed on the water bath and the sample concentrated to

2 mL. The Snyder column is removed and the volume is brought up to 5 mL with methylene

chloride. The extract is placed in a vial with minimal headspace, the solvent level is carefully,

marked, the vial is labeled, and placed in storage at 4°C until analysis.

Sample extracts will be analyzed for semivolatile organic compounds using the analytical

procedures of SW-846 Method 8270. Instrument operating conditions are shown in

Method 8270 and the laboratory standard operating procedure. The mass spectrometer will be

tuned and mass calibrated as required using peffluorotributylamine (FC-43), as per the

manufacturer's instructions. The tune of the instrument is verified by injecting 50 ng of

decafluorotriphenylphosphine (DFTPP) and checking the ion abundance criteria against the ion

abundance criteria listed in Method 8270. If the DFTPP mass spectrum does not meet method

specifications, the DFTPP is re-analyzed or the mass spectrometer is re-tuned so that the

instrument will meet the tuning criteria. The DFTPP tune criteria must be met before analysis of

samples can begin. The acceptability of the instrument tune will be verified by analysis of the

DFTPP solution every twelve hours.

Method 8270 calibration procedures and criteria apply. Calibration check compounds

and system performance check compounds must meet the criteria outlined in Method 8270. A 9

multipoint calibration is performed initially to determine system response factors for the

compounds of interest; system calibration is verified daily (or every 12 hours) by analyzing a

mid-level calibration standard in accordance with Method 8270 specifications. All samples will

be spiked with Method 8270 internal standards immediately prior to analysis.

A solvent blank is analyzed prior to sample analysis to demonstrate that the analytical

system is free from contamination. Internal standard area counts for each sample analysis must

be between 50 and 150% of the last daily calibration standard, in accordance with Method 8270

specifications. Surrogate compound recoveries1 for each sample are checked against

Method 8270 surrogate spike recovery limits for soil/sediment samples. - Surrogate compound

glp/D:\SECT8.WPD

Project No. 0121.00 • Element No. B3 Revision No. 1 Date March 2000 Page 13 of 17

recoveries from the XAD-2® and the sample matrix used in this work have not been established

experimentally; recoveries of the listed compounds from XAD-2® will be determined

experimentally prior to the initiation of the monitoring program. Samples will be re-analyzed if

surrogate compound recoveries fall below the levels specified in Method 8270. In order to

evaluate recoveries of compounds of interest from XAD-2®, a method detection limit study will

be performed using spiked XAD-2® and the procedures of 40 CFR Part 136B.. The spiked

samples will be subjected to the same sample preparation procedures as the samples generated in

the field.

Criteria for identification of the mass spectra of the compounds of interest are positive

matching of the relative retention times and the mass spectra of the sample and the standard

components in accordance with the specifications of Method 8270. Quantitative analysis is

achieved by the use of automated procedures in'the Hewlett-Packard data system. Mass spectral

interpretation of tentatively identified compounds (if performed) will be verified manually by

experienced interpreters of mass spectral data using the NBS reference library, with automated

semi-quantitative analysis achieved by comparison of the peak area of the tentatively-identified

compound to that of the closest-eluting standard.

The compounds of interest and iexperimental method detection limits are shown in

Table. 8-2.

8.7 Ethylene Oxide by Gas Chromatograph Analytical Systems

Ethylene Oxide samples are stored in the refrigerator after they are received from the field

prior to analysis. Sample preparation is performed by scoring the sample tube with a file and

glp/D:\SECT8.WPD

Project No. 0121.00 Element No. B3 Revision No. 1 Date March 2000 Page 14 of 17

Table 8-2

Semivolatile Organic Compounds to be Analyzed by the Analytical Procedures of Method 8270; with Method Detection Limits

Compound MDL

0*gj Compound MDL > g )

acenaphthene. 5.82 2-methylphenol 9.44

acenaphthylene 8.87 4-methylphenol 7.69

acetophenone 13.87 naphthalene 15.46

aniline 16.20 1-naphthylamine 5.42

anthracene 17.06 2-naphthylamine 10.22

4-aminobiphenyl 9.60 2-nitroaniline 12.30

benzidine 0.00 3-nitroaniline 8.70

benzoic acid 12.3d 4-nitroaniline 10.40

benzo(a)anthracene 8.33 nitrobenzene 24.86

benzo(b)fluoranthene 17.34 2-nitrophenol 10.07

benzo(k)fluoranthene 23.61 . 4-nitrophenol 7.36

benzo(g,h,i)perylene 15.05 N-nitroso-di-n-butylamine 22.42

benzo(a)pyrene 18.17 N-nitrosodiphenylamine 50

benzyl alcohol 8.15 N-nitrosodipropylamine 21.48

bis(2-chloroethoxy)methane 14.03 N-nitrosopiperidine 17.32

bis(2-chloroethyl)ether 11.66 pentachlorobenzene 9.74

bis(2-chloroisopropyl)ether 11.07 pentachloronitrobenzene 10.37

bis(2-ethylhexyl) phthalate 11.62 pentachlorophenol 14.74

4-bromophenyl phenyl ether 11.30 phenanthrene 10.29

butyl benzyl phthalate 11.66 phenol 22.48

4-chloroaniline 16.83 2-picoline 11.15

glp/D:\SECT8.WPO

Project No. Element No. Revision No. Date Page

Table 8-2

(Continued)

Compound MDL (Hg) Compound

MDL (^g)

1 -chloronaphthalene 30.03 pyrene 10.54

2-chloronaphthalene 18.48 1,2,4,5-tetrachlorobenzene 10.54

4-chloro-3-methylphenol 16.80 2,3,41,6-tetrachlorophenol 9.67

2-chlorophenol 9.99, 1,2,4-trichlorobenzene 13.62

4-chlorophenyl phenyl ether 6.79 2,4,5krichlorophenol 6.90

chrysene 10.57 2,4,6-trichlorophenol 8.35

dibenz(a,h)anthracene j 15.85 diphenylamine 25.39

dibenzofuran \ 9.28 1,2-diphenylhydrazine 50

di-/z-butyl phthalate 14.01 ' dir«-octyl phthalate 13.19

1,3-dichlorobenzene 12.05 fluoranthene 14.49

1,4-dichlorobenzene 10.86 fluorene 10.01

1,2-dichlorobenzene t 10.96 hexachlorobenzene 14,12

3,3'-dichlorobenzidine \ ' 8.95 hexachlorobutadi ene 13.23

2,4-dichloro phenol , 14.55 hexachlorocyclopentadiene 21.74

2,6-dichlorophenol 18.10 hexachloroethane 5.65

diethyl phthalate 7.20: inderio( 1,2,3 -cd)pyrene 14.56

p-dimethylaminoazobenzene 50 !: isophorone 22.61

7,12-dimethylbenz(a)anthracene ; 19.04 methyl methanes ulfonate 16.50

a-, a-dimethylphenethylamine 1 oov 2-methylnaphthalene 11.36

2,4-dimethylphenol 17.64 2,4-dinitrophenol 10.05

dimethyl phthalate 9.15 2,4-dinitrotoluene 9.71

4,6-dinitro-2-methylphenol 11.31 2,6-dinitro toluene 9.38

0121.00 B3

1 March 2000

15 of 17

glp/D:\SECT8.WPD

Project No. 0121.00 Element No. B3 Revision No. 1 Date March 2000 Page 16 of 17

Table 8-2

(Continued)

removing the front sorbent section and first glass wool plug to a 2 mL vial. The middle glass

wool plug and back sorbent section goes into another vial. One milliliter of

dimethylformamide(DMF) is added to each vial and they are capped. Each sample is shaken for

10 seconds and allowed to stand for 5 minutes. A 20 pL aliquot of the DMF solution is

transferred to another 5 mL vial containing 2.0 mL 2% N-heptafluorobutyrylimidazole (HFB1)

(v/v) in isooctane. Each sample is shaken for 1 minute to ensure complete hydrolysis of excess

HFB1 forming 2-bromoethylheptafluorobutyrate. One milliliter of this sample is transferred to a

2 mL vial and analyzed by GC/ECD.

The analysis of 2-bromoethylheptafluorobutyrate is performed using a Varian® 9000 GC

equipped with an ECD detector. This system is configured with a 3m by 4 mm glass column

with 10% SP-100 on 80/100 Chromosorb WHP; ERG's system uses a Nelson Analytical

A/D interface as the data system. Typically, 1-uL samples are injected with an automatic sample

injector. The carrier gas composition for the analysis is 5% methane in argon at 25 mL/minute.

The detection range for the ethylene oxide analysis is 2 to 42 pg ethylene oxide per sample.

8.8 DioxinVFuran by High Resolution Mass Spectroscopy Analytical Systems

After receipt of the sample shipment, the samples are checked against the Chain-of-

Custody forms and then assigned an analytical laboratory sample number. Each sample

glp/D:\SECT8.WPD

Project No. 0121.00 Element No. B3 Revision No. 1 Date March 2000 Page 17 of 17

component is examined to determine if damage occurred during travel. Color, appearance, and

other particulars of the samples are noted. Samples are extracted within 21 days of collection

and processed through cleanup procedures before concentration and analysis.

The analytical procedure used to obtain PCDD/PCDF concentrations uses high resolution

gas chromatography (HRGC) coupled with high resolution mass spectrometry (HRMS), with a

resolution from 8000-10000. Dioxin 2,3,7,8-tetrachlorpdibenzo-p-dioxin is to be analyzed by

Columbia Laboratories, Inc., using Compendium Method TO-9A and Method 8290.

In case of a failure in the analytical system, Columbia Laboratories will be responsible

for corrective action.

8.9 Metals Using an Inductively Coupled Argon Plasma Mass Spectroscopy Analytical System

After receipt of the sample shipment, the samples are checked against the Chain-of-

Custody forms and then assigned an analytical laboratory sample number. Each sample

component is examined to determine if damage occurred during travel. Color, appearance, and

other particulars of the samples are noted. Samples are extracted within according to standard

procedures determined by First Analytical.

The analytical procedure used to obtain the metal concentrations uses inductively coupled

argon plasma mass spectrometry (ICPMS). The metals are to be analyzed by First Analytical,

using approved methods as published in 40 CFR Parts 50, 51, and 58.

glp/D:\SECT8.WPD

Project No. Element No. Revision No. Date

0121.00 B4

1 March 2000

1 of 16 Page

SECTION 9

QUALITY CONTROL REQUIREMENTS

This section describes the quality control requirements performed under each of the major

program components (NMOC, SNMOC, UATMP, Carbonyls, PAMS, and HAPS).

9.1 . Sample Canister Cleanup Studies

Before any samples are collected for a program, all stainless steel sample canisters are

checked for leaks. The canisters are filled to about 15 psig pressure with zero-grade air from a

cylinder and the valve.and fittings are checked for leaks. The canisters are then cleaned using the

procedure described in Section 9.

After cleanup, each canister is analyzed and the results are noted on the custody form for

the permanent record. In order for the canister to be used without further cleanup, the analysis

must show that it meets the quality objective for cleanliness.

9.2 Standard Traceability

The standards used for the NMOC/SNMOC and PAMS are vendor-supplied National

Institute of Standards and Technology (NIST) standards or referenced to a vendor-supplied NIST

standard. The standards used for UATMP are certified by comparison to external audit samples.

The SOP used to prepare standards is the SOP for Standard Preparation Using Dynamic Flow

Dilution, ERG-MOR-061.

glp/D:\SECT9.WPD

Project No. Element No. Revision No. Date

0121.00 B4

1 March 2000

2 of 16 Page

9.3 Accuracy and Acceptance

Because ambient air measurements encompass a range of mixtures of organic compounds

whose individual concentrations are unknown, defining absolute accuracy is not possible.

Instead, accuracy is determined relative to standards with internal and external audit samples.

9.3.1 NMOC Instrument Calibration

Accuracy is monitored throughout the program using in-hduse QC samples. On days

when ambient air samples are analyzed for NMOC content, an in-house QC sample of propane is

also analyzed. The QC samples are prepared by diluting dry propane with cleaned, dried air

using calibrated flowmeters. The concentration of the in-house QC samples will be set close to

the concentration levels seen in the ambient air samples.

9.3.2 SNMOC Instalment Calibration

Because all samples analyzed for volatile analyses utilize the same instrument and have

the potential to report all analytes possible, the hydrocarbon and TO-14A/TO-15 parameters

must pass the standard procedures' set; and listed in Tables 4-2 and 4-3.

Prior to sample analysis for SNMOC, a quality control standard, prepared using either a

Scott Specialty Gas or Spectra certified high pressure gas, is analyzed daily to ensure the

validity of the current monthly response factors. This standard will have an approximate range

from 15 ppbv to 40 ppbv.

For each detector, load volumes and the standard response area counts are entered into a

computer spreadsheet and the current monthly response factors are used to calculate propane

concentrations. The concentration is compared to the calculated theoretical concentrations of the

glp/D:\SECT9.WPD

Project No. Element No. Revision No. Date

0121.00 B4

1 March 2000

3 of 16 Page

QC standard. A percent bias less than or equal' to 30 percent for each compound is considered

acceptable.

If the QC standard does not meet the 30 percent criterion, a second QC standard is

analyzed. If the second QC standard meets the criterion, the analytical system is considered in

control. If the second QC check does not meet'acceptance criteria, a leak test and system

maintenance are performed. Following this, a third QC standard analysis can be performed. I f

the criterion is met by the third analysis, the analytical system is considered in control. I f

maintenance causes a change in system response, a new calibration curve is required.

A system blank of cleaned, humidified air is analyzed after the daily QC standard

analysis before the sample analysis. The system is considered in control i f the total NMOC

concentration for the system blank is less than or equal to 20 ppbC.

- 9.3.3 UATMP Instrument Calibration

The tune of the MSD is verified using 4-bromofluorobenzene (BFB) on a daily basis.

The tune is usually verified during the analysis of the QC sample. Before sample analyses, a

standard prepared at approximately 5 ppbv from a certified cylinder is used for a daily

calibration. The resulting response factors for each compound will be compared to the monthly

calibration curve response factors generated from the GC/MSD using the HP Chemstatioh®

Software. An absolute value of less than or equal to 30 percent is considered acceptable for the

quantitated compounds. If the first QC check does not meet this criterion, a second standard may

be analyzed. If the second QC standard is acceptable, sample analysis can continue. I f the

second check does not meet acceptance criteria, a leak check and system maintenance are •i

performed. If the system maintenance is completed and a third QC analysis meets the criterion,

then analysis may continue. If the maintenance causes a change in the system response, a new

calibration curve must be analyzed before sample analyses can continue.

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 4 of 16

After acceptance of the daily standard, a system blank consisting of cleaned, humidified

air is analyzed. A total concentration of less than 0.2 ppbv (or the lowest quantitation limit,

whichever is smaller) per compound indicates that the system is in control. I f a concentration

greater than the acceptance criterion is detected, a second system blank is analyzed. If the second

system blank fails, system maintenance is performed. Another system blank can be analyzed and

if it is in control, ambient air samples are analyzed.

9.3.4 PAMS VOC Analysis

Daily QC checks for PAMS VOC analysis are the same as those described for SNMOC

in Section 9.3.2.

9.3.5 Carbonyl Compounds Analysis

Daily calibration checks are performed to ensure that the analytical procedures are in

control. Daily QC checks are performed after every 10 samples on the days that samples are

analyzed. Compound responses must be within 15% bias relative to the responses from the

current calibration curve. Compound retention time drifts are also measured from this analysis

and tracked to ensure that the HPLC instrument is operating within acceptable parameters.

If this daily QC check does not meet the criterion, a second injection of the QC standard

is performed. If the second QC check does notpass or i f more than one daily QC check does not

meet the criterion, a new calibration curve (5 concentration levels) is analyzed. All samples

analyzed with the unacceptable QC check will be reanalyzed.

An acetonitrile system blank is analyzed after the daily calibration check and before

sample analysis. The system is considered in control if target compound concentrations are less

than the current detection limits.

glp/D:\SECT9.WPD

Project No. Element No. Revision No Date Page

9.3.6 Analysis of Semivolatile Organic Compounds Using EPA Method 8270

Prior to using Method 8270, the filter/sorbent samples are prepared for analysis using the

procedures of EPA Method 3542 (Appendix I). The extracts are analyzed by GC/MS, using a

fused silica capillary column and a mass spectrometer capable of scanning from 35 to 500 mass

units every 1 sec or less, using a nominal electron energy of 70 eV in the electron ionization

mode. The mass spectrometer must be capable of producing a mass spectrum for

decafluorotriphenylphosphine (DFTPP) that meets all of the acceptance criteria in Table 9-1

when 1 pL of the GC/MS tuning standard is injected through the GC.

Table 9-1

Decafluorotriphenylphosphine (DFTPP) Key Ions and Ion Abundance Criteria According to EPA Method 8270

Mass Ion Abundance Criteria

51 30-60% of mass 198

68 <2%ofmass69

70 <2%ofmass69

127 40-60% of mass 198

197 <1% of mass 198

198 Base peak, 100% relative abundance

199 5-9% of mass 198

275 ' 10-30% of mass 198

365 >1% of mass 198

441 Present but less than mass 443

442 >40%ofmass 198

443 .17-23% of mass 442 ' *

0121.00 B4

1 March 2000

5 of 16

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. • B4 Revision No. 1 Date March 2000 Page 6 of 16

The Method 8270 surrogate compounds will be used to spike the sorbent immediately

before extraction. These surrogate compounds are phenol-d6, 2-fluorophenol,

2,4,6-tribromophenol, nitrobenzene-d5, 2-fluorobiphenyl, and /?-terphenyl-d14. Surrogate

recovery ranges have not been established for XAD-2® extraction; surrogate recovery acceptance

ranges from Method 8270 will be used as a guideline.

A GC/MS system performance check must be performed to ensure that minimum average

response factors are met before the multipoint calibration is used. For semivolatile organic

compounds, the System Performance Check Compounds are N-nitroso-di-«-propylamine,

hexachlorocyclopentadiene, 2,4-dinitrophenol, and 4-nitrophenol. These System Performance

Check Compounds typically have very low response factors (0.1 - 0.2) and the response factors

tend to decrease as the chromatographic system begins to deteriorate or as the standard begins to

deteriorate. These compounds are usually the first to show poor performance, and these

compounds must, therefore, meet the minimum requirement when the system is calibrated.

After the analytical system performance check is met, the calibration check compounds

are used to check the validity of the initial multipoint calibration. These calibration check

compounds are acenaphthene, 1,4-dichlorobenzene, hexachlorobutadiene, N-nitroso-di-

phenylamine, di-«-octyl phthalate, fluoranthene, benzo(a)pyrene, 4-chloro-3-methylphenol,

2,4-dichlorophenol, 2-nitrophenol, phenol, pentachlorophenol, and 2,4,6-trichlorophenol. The

response factor for the calibration check compounds must be within ±30% of the mean response

factor from the initial calibration.

Internal standard responses and retention times must also be evaluated for stability. EPA

Method 8270 also presents detailed guidelines for qualitative analysis of mass spectra, as well as

a detailed analytical scheme to determine that all target analytes are quantitated relative to the

nearest-eluting internal standard.

9.3.7 Ethylene Oxide Analysis

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 7 of 16

Daily calibrations with at least six working standards over the range of 1 to 42 pg

ethylene oxide per sample is performed to ensure that the analytical procedures are in control.

Daily QC checks are performed after every lO samples with response within 20% bias relative to

the responses from the calibration curve. Compound retention time drifts are also measured from

this analysis and tracked to ensure that the instrument is operating within acceptable parameters.

If this daily QC check does not meet the criterion, a second injection of the QC standard

is performed. If the second QC check does not pass or if more than one daily QC check does not

meet the criterion, a new calibration curve (6 concentration levels) is analyzed. All samples

analyzed with the unaccepted QC check will be reanalyzed. Three system blanks are prepared

and analyzed to ensure that the calibration graph is in control.

9.3.8 Analysis of Polychlorinated Dibenzodioxins Using EPA Compendium Method TO-9 and EPA Method 8290 .,

EPA Method 8290 provides procedures for the determination of polychlorinated dibenzo-

/7-dioxins (tetra- through octachlorinated dioxins) in a variety of environmental matrices at part-

per-trillion to part-per-quadrillion concentrations. The Method 8290 analytical methodology

requires the use of high resolution gas chromatography (HRGC) coupled with high resolution

mass spectrometry (HRMS). Method 8290 has been applied to water, soil, sediment, paper pulp,

fly ash, fish tissue, human adipose tissue, sludges, and still fuel oil but the sample preparation

and purification methodology of Method 8290 is not directly applicable to ambient air samples

collected on PUF according to EPA Compendium Method TO-9A. The analytical procedures of

EPA Method 8290, however, are directly applicable to the ambient samples generated by EPA

Method TO-9A.

A Soxhlet extraction using toluene is used to extract the dioxins/furans from the filter and

PUF sampling module. Ambient air samples do not generally require the extensive chemical

cleanup procedures described in Method 8290. Toluene extracts are concentrated and analyzed

by HRGC/HRMS.

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 8 of 16

The HRGC/HRMS system is calibrated using isotopically labeled internal standards and

recovery standards, with a number of unlabeled analytes that are representative of the various

chlorine number congeners.

Prior to calibration, the HRMS is tuned to the manufacturer's specifications with a

minimum resolving power of 10,000 using perfluorokerosene. The lowest point of the

calibration is near but not at the method detection limit, and the upper point of the calibration is

adjusted to cover the anticipated concentration range of the samples. The calibration is

acceptable if the percent relative standard deviations for the mean response factors from the

17 unlabeled standards do not exceed ±20%, and those for the nine labeled reference compounds

do not exceed ±30%.

A'column performance check solution is used to verify that the performance of the

HRGC column meets method specifications. Because Method 8290 is based upon the

application of Selected Ion Monitoring techniques to identify and quantify the analytes, method

specifications for the required abundance ratios of chlorine isotopes at the different chlorine

numbers are presented. If at least two peaks of the chlorine isotope cluster at a given chlorine

number (i.e., pentachlorodibenzodioxin) are not present in the correct ratio relative to the

theoretical value, the compound cannot be identified as pentachlorodibenzodioxin.

Because additional air samples will not be available for preparation/analysis of a matrix

spike/matrix spike duplicate, a method spike/method spike duplicate will be prepared and

analyzed in the laboratory using clean sorbent media.

9.3.8 Quality Control Measures for Analysis of Airborne Metals Collected on a Filter

Analysis of the metals will be performed by inductively coupled argon plama mass

spectroscopy for antimony, arsenic, beryllium, cadmium, total chromium, lead, manganese and

glp/D:\SECT9.WPD

Project No. Element No. Revision No. Date

0121.00 -B4

1 March 2000

9 of 16 Page

nickel, and by cold vapor atomic absorption spectrophotometry for mercury. Quality control

measures for the metals analysis are shown in Table 9-2.

9.4 Precision

Analytical precision is estimated by repeated analysis of samples. For all samples, the

second analysis is performed at least 24 hours after the first analysis. This procedure is followed

for the canister samples in particular to ensure that sufficient time has elapsed to allow the

canister contents to equilibrate with the solid surfaces and to allow any concentration gradients

within the canister to disperse.

Duplicate samples are reanalyzed once each to determine overall precision, including

sampling and analysis variability.

Precision estimates are calculated in terms of percent difference and absolute percent

difference. Because the true concentration of the ambient air sample is unknown, these

calculations are relative to the average sample concentration.

9.5 Completeness

Completeness, a quality measure, is calculated at the end of the program. Percent

completeness is calculated as the ratio of the number of valid samples received to the number of

scheduled samples (beginning with the first valid sample received through the last sample

received). This quality measure is presented in the final report.

glp/D:\SECT9.WPD

Table 9-2

Quality Control Measures for Metals Analysis

Parameter Frequency Acceptance Criteria Corrective Action

Multipoint calibration Daily Correlation coefficient > 0.995

1. Repeat analysis of calibration standards 2. Re-prepare calibration standards and re-analyze

Calibration check Daily Recovery 95-105% for all analytes

1. Repeat analysis of calibration check standard 2. Repeat analysis of calibration standards 3. Re-prepare calibration standards and re-analyze

Continuing calibration verification

Every 10 samples Recovery 90-110% 1. Repeat analysis of continuing calibration verification sample 2. Reprepare continuing calibration verification sample and re-analyze 3. Reanalyze samples since last acceptable continuing calibration verification

Method blanks Every 10 samples Analytes below method detection limit

1. Reanalyze 2. Reprepare blank and re-analyze 3. Correct contamination and reanalyze blank 4. Repeat analyses of all samples since last clean blank

Laboratory control sample One per sample batch Recovery 80-120% Reprepare sample batch; re-analyze

Method spike/method spike duplicate

one per sample batch Recoveries 80-120% Flag data.

001210 0003 TOW

T vu 9U<*01

Project No. 0121.00 v Element No. B4

* Revision No. 1 Date March 2000

i Page 11 of 16

9.6 Representativeness

Representativeness measures how well the reported results reflect the actual ambient air

concentrations. This measure of quality can be enhanced by ensuring that a representative

sampling design is employed. This design includes proper integration over the desired sampling

period and following siting criteria established for each task. The experimental design for

sample collection should ensure samples are collected at proper times and intervals for their

designated purpose according to the data quality objectives. For example, NMOC/SNMOC

samples are collected to gain information about PAMS VOCs. Therefore, collection of 3-hour

samples from 6:00 a.m. to 9:00 a.m. each weekday is appropriate. Quality measures of duplicate

sample collection and replicate analyses are included.

9.7 Comparability

Comparability is a measure of how well the program data compare to like data. Sample

exchange is a means of determining comparability.

When EPA directs, an exchange of NMOC samples can be made with EPA-QAD and the

EPA's National Exposure Research Laboratory (NERL). SNMOC and PAMS samples can be

exchanged with EPA-NERL for concentration;comparisons and with GC/MSD for identification

confirmations. For sites choosing site support̂ an exchange of samples between the site's

analytical laboratory and ERG can be conducted. i

• t (

9.8 Lowest Quantitation Limits

For SNMOC, UATMP and carbonyls,"the lowest quantitation limits of the target

compounds are determined by performing seven replicate analyses of a standard that is at a

concentration within five times the expected quantitation limits. This procedure follows the

method listed in the Federal Register, Appendix B, Part 136.(17) The quantitation limits

glp/D:\SECT9.WPD

Project No. Element No. Revision No. Date

0121.00 B4

1 March 2000

12 of 16 Page

determined are verified by analyzing multiple injections of standards at the obtained limits to

confirm the reported lowest level concentration.:

The lowest quantitation limits for the SNMOC are listed in Table 9-3, for UATMP

compounds in Table 9-4, and for the carbonyl compounds in Table 9-5. All laboratories at

ERG's Morrisville location verify the lowest quantitation limits once a year by preparing and

analyzing the seven replicate standards. The semivolatile quantitation limits were presented

previously in Table 9-1. Dioxin and metal detection limits are summarized in the separate

subcontractor quality assurance plans.

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 13 of 16

Table 9-3

SNMOC Lowest Quantitation Limits

Compound Quantitation Limit (ppbC) Compound

Quantitation Limit (ppbC)

Acetylene/Ethane 0.62 . ; Meth'ylcyclohexane 3.72

Benzene 2.13 Methylcyclopentane 2.13

1,3-Butadiene 0.20 2-Methylheptane 4.73

«-Butane 0.20 3-Methylheptane 4.73

cz's-2-Butene 0.20 2-Methylhexane 3.72

/ra«5-2-Butene 0.20 3-Methylhexane 3.72 '

Cyclohexane 2.13 2-Methylpentane 2.13

Cyclopentane 0.37 3-Methylpentane 2.13

Cyclopentene 0.37 2-Methyl-1-Pentene 2.13

/7-Decane 4.60 4-Methyl-1-Pentene 2.13

1-Decene 4.60 /7-Nonane 4.60

m-Diethylbenzene 4.60 1-Nonene 4.60

/?-Diefhylbenzene 4.60 n-Octane 4.73

2,2 -Dimethy lbutane 2.13 1-Octene 4.73

2,3-Dimethylbutane 2.13 rc-Pentane 0.37

2,3-Dimethylpentane 3.72 1 -Peritene 0.37

2,4-Dimethylpentane 3.72 . cw-2-Pentene 0.37

n-Dodecane 4.60 trans-2-Venter\Q 0.37

1 -Dodecene 4.60 a-Pinene 4.60

2-Ethyl-l-Butene 2.13 P-Pinene 4.60

Ethylbenzene 4.73 Propane 0.31

Ethylene 0.62 «-Propylbenzene 4.60

glp/D:\SECT9.WPD

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 14 of 16

Table 9-3

(Continued)

Compound Quantitation Limit (ppbC) , Compound

Quantitation Limit (ppbC)

m-Ethyltoluene 4.60 Propylene 0.31

o-Ethyltoluene 4.60 Propyne 0.31

p-Ethyltoluene 4.60 Styrene 4.73

//-Heptane 3.72 Toluene 3.72

1 -Heptene 3.72 «-Tridecane 4.60

«-Hexane 2.13 1 -Tridecene 4.60

1 -Hexene 2.13 "• 1,2,3-Trimethylbenzene 4.60

c/.s-2-Hexene 2.13 ' 1,2,4-Trimethylbenzene 4.60

trans-2-Hexene 2.13 1,3,5-Trimethylbenzene 4.60

Isobutane 0.20 2,2,3 -Trimethylpentane 4.73

Isobutene/1-Butene 0.20 . 2,2,4-Trimethylpentane 4.73

Isopentane 0.37 ; 2,3,4-Trimethylpentane 4.73

Isoprene 0.37 «-Undecane '4.60

Isopropylbenzene 4.60 1-Undecene 4.60

2-Methyl-1-Butene 0.37 . m.jC-Xylene 4.73

2-Methyl-2-Butene 0.37 o-Xylene 4.73

3-Methyl-1-Butene • 0.37

Project No. 0121.00 Element No. B4 Revision No. 1 Date March 2000 Page 15 of 16

Table 9-4

TO-15 Analyte Lowest Quantitation Limit (LQL)

Compound ppbv Compound ppbv

Acetylene 0.10 1 ;2-Dichloropropane 0.04 Propylene 0.10 Ethyl Acrylate 0.10 Chloromethane 0.13 Bromodichloromethane 0.05 Vinyl Chloride 0.06 : Trichloroethylene 0.04 1,3-Butadiene 0.09 "i

i

Methyl Methacrylate 0.07 Bromomethane . 0.1.4 c/.s-l,3-Dichloropropene 0.06 Chloroethane , 0.06 Methyl Isobutyl Ketone 0.07 Acetonitrile 0.57 ?ra«5-l,3-Dichloropropene 0.11 Acrylonitrile 0.21 1,1,2-Tri chloroethane 0.11 Methylene Chloride 0.09 Toluene 0.21 trans-1,2-Dichloroethylene 0.12 Dibromochloromethane 0.15 1,1 ,-Dichloroethane 0.10 ; «-Octane 0.21 Methyl tert-Butyl Ether 0.06 1 Tetrachloroethylene 0.22 Methyl Ethyl Ketone 0.17 Chlorobenzene 0.07 Chloroprene ; 0.10 Ethylbenzene 0.12 Bromochloromethane 0.09 m,p-Xy\ene 0.23 Chloroform 0.06 . ' Bromoform 0.15 Ethyl tert-Butyl Ether : 0.07 ! Styrene 0.10 1,2-Dichloroethane 0.06 1,1,2,2-Tetrachloroethane 0.17 1,1,1 -Trichloroethane 0.05 o-Xylene 0.10 Benzene 0.07 m-Dichlorobenzene 0.15 Carbon Tetrachloride 0.05 p-Dichlorobenzene 0.13 tert-Amyl Methyl Ether 0.07 o-Dichlorobenzene 0.16

glp/D:\SECT9.WPD

•

m O

Table 9-5

Carbonyl Lowest Quantitation Limits, Underivatized Concentration (ppbv)

COMPOUND SAMPLE VOLUME (L)

100 200 300 400 500 600 700 800 900 1000

Formaldehyde 0.03 0.02 0.011 0.008 0.006 0.005 0.005 0.004 0.004 0.003 Acetaldehyde 0.04 0.02 0.013 0.010 0.008 0.007 0.006 ' 0.005 0.004 . 0.004 Acrolein 0.05 0.02 0.016 0.012 0.010 0.008 0.007 0.006 0.005 0.005 Acetone 0.03 0.01 0.009 0.007 0.005 0.004 0.004 0.003 0.003 0.003 Propionaldehyde 0.02 0.01 0.006 0.005 0.004 0.003 0.003 0.002 0.002 0.002 Crotonaldehyde 0.04 0.02 0.013 0.010 0.008 0.006 0.005 0.005 0.004 0.004 Butyr/lsobntyraldehyde 0.05 0.02 0.015 0 011 0.009 0.008 0.006 0.006 0.005 0.005 Benzaldehyde 0.02 0.01 0.008 0.006 0.005 0.004 0.003 0.003 0.003 0.002 Isovaleraldehyde 0.10 0.05 0.034 0.025 0.020 0.017 0.014 0.013 0.011 0.010 Valeraldehyde 0.06 0.03 0.021 0.016 0.013 0.011 0.009 0.008 0.007 0.006 Tolualdehydes 0.09 0.05 0.031 0̂ 023 0.019 6.016 0.013 0.012 0.010 0.009 Hexaldehyde 0.04 0.02 0.013 0.010 0.008 0.006 0.006 0.005 0.004 0.004 2,5 -dimethy lbenzaldehy de 0.05 0.03 0.017 0.013 0.010 0.008 0.007 0.006 0.006 0.005

COMPOUND SAMPLE VOLUME (L)

1100 1200 1300 1400 1500 1600 1700 1800 1900 2000

Formaldehyde 0.0029 0.0026 0.0024 0.0023 0.0021 0.0020 0.0019 0.0018 0.0017 0.0016 Acetaldehyde 0.0036 0.0033 0.0030 0.0028 0.0026 0.0025 0.0023 0.0022 0.0021 0.0020 Acrolein - 0.0044 0.0040 0.0037 0.0035 0.0032 0.0030 0.0029 0.0027 0.0026 0.0024 Acetone 0.0024 0.0022 0.0020 0.0019 0.0018 0.0017 0.0016 0.0015 0.0014 0.0013 Propionaldehyde 0.0017 0.0016 0.0015 0.0014 0.0013 0.0012 0.0011 0.0011 0.0010 0.0010 Crotonaldehyde 0.0035 0.0032 0.0029 0.0027 0.0025 0.0024 0.0022 0.0021 0.0020 0.0019 Butyr/Isobutyraldehyde 0.0041 0.0038 0.0035 0.0032 0.0030 0.0028 0.0027 0.0025 0.0024 0.0023 Benzaldehyde 0.0022 0.0020 0.0018 0.0017 0.0016 0.0015 0.0014 0.0013 0.0012 0.0012 Isovaleraldehyde 0.0092 0.0084 0.0078 0.0072 0.0067 0.0063 0.0059 0.0056 0.0053 0.0050 Valeraldehyde 0.0058 0.0054 0.0049 0.0046 0.0043 0.0040 0.0038 0.0036 0.0034 0.0032 Tolualdehydes 0.0085 0.0078 0.0072 0.0067 0.0063 0.0059 0.0055 0.0052 0.0049 0.0047 Hexaldehyde 0.0035 0.0032 0.0030 0.0028 0.0026 0.0024 0.0023 0.0021 0.0020 0.0019 2.5-dimethvlbenzaldehvde 0.0046 0.0042 0.0039 0.0036 0.0034 0.0032 0.0030 0.0028 0.0027 0.0025

CTQ CO

d CD < cT

o

Z O

>-t

a o

~ Z Z o o •

o rr o

Project No. 0121.00 Element No. B5 Revision No. 1 Date March 2000 Page 1 ol"2

SECTION 10

INSTRUMENT/EQUIPMENT TESTING, INSPECTION, AND MAINTENANCE REQUIREMENTS

To ensure the quality of the sampling and analytical equipment, ERG conducts

performance checks for all equipment used in the programs. ERG personnel check, and if

needed, repair the sampling systems before the seasons begin each year. ERG tracks the

performance of the GCs to ensure proper operation. ERG also maintains a spare parts inventory

to prevent equipment downtime.

10.1 NMOC

The Hewlett-Packard (H-P) GCs used for NMOC measurements are maintained on an

as-needed basis. Before the beginning of the analytical season, an H-P technical service

representative performs preventive maintenance. Throughout the analytical season, minor

maintenance is performed by ERG personnel.

The SNMOC analytical system is maintained as described in Section 10.2.

10.2 SNMOC, UATMP, and PAMS

The GC/FID/MSD system is maintained under a service agreement. Twice a year,

preventive maintenance is performed by a technical representative. ERG personnel perform

minor maintenance, such as ferrule changes, carrier gas filter replacements, column maintenance,

and source cleaning. ,

glp/D:\SECT10.WPD

Project No. 0121.00 Element No. B5 Revision No. 1 Date March 2000 Page 2 of 2

10.3 PAMS

The VOC PAMS analytical system is maintained as described in Section 10.2.

The PAMS carbonyl HPLC analytical system receives preventive maintenance by a

technical service representative before the beginning of the analytical season. ERG personnel

perform other minor maintenance, such as column and detector maintenance, on an as-needed

basis.

10.4 HAPS

The GC/MSD system is maintained under a service agreement! Twice a year, preventive

maintenance is performed by a technical representative. ERG personnel perform minor

maintenance, such as ferrule changes, carrier gas filter replacements, column maintenance, and

source cleaning.

For the other HAPs sample analyses performed on the GC/ECD,' HRMSD, and ICPMS

analytical systems, preventive maintenance is performed by competent technical service

representatives as needed. ERG and subcontractor personnel perform minor maintenance, such

as column and detector maintenance, on an as-needed basis.

glp/D:\SECT10.WPD

Project No. Element No. Revision No. Date

0121.00 B6

1 March 2000

1 of 11 Page

SECTION 11

INSTRUMENT CALIBRATION AND FREQUENCY

Because the requirements of the programs for analytical system calibrations differ, the

programs are discussed separately in this section.

11.1 NMOC Calibration

Before the analytical program begins, an NMOC calibration curve is generated at the

beginning and the end of the sampling season using five propane NIST certified standards and

zero air. Propane concentrations for the calibration curve are prepared at a concentration range

from zero to 10 ppmC. Zero concentration air is made from clean humidified air, The standards

are prepared from NIST certified cylinders. These standards are analyzed directly from an NIST

certified standard into the GC-PID.

Calibration curves are calculated by linear regression, assuming a linear relationship

between area counts and concentration. If the regression coefficient for any channel is less than

0.995, the entire curve is regenerated.̂ If a relative standard deviation of 3% for each point is not

met, the point is repeated.

Response factors for the NMOC calibration are verified every morning samples are

analyzed by making two injections of the mid-range QC calibration standard, an independently

prepared calibration standard with a concentration of approximately 3.0 ppmC. The relative

standard deviation (RSD) is computed and a third injection is analyzed i f the RSD is greater than

3%. After the third injection, the RSD is computed again.

RSD = SD

(Eq. 114) SamDle Averaee

glp/D:\SECT11.WPD

Project No. 0121.00 Element No. B6 Revision No. 1 Date March 2000 Page 2 of 11

where:

SD = Sample Standard Deviation calculated with the denominator of N- l . N = Number of injections. •

The relative error from the pair of QC injections should be within 20% of the theoretical

concentration.

TPC - DPC Relative Error = — * 100 (Eq. 11-2)

TPC

where:

TPC = Theoretical Propane Concentration. DPC = Daily Propane Concentration.

If the QC value does not meet the 20%'requirement, the QC check analysis is repeated. I f

the 20% requirement is again not met, the analysis is repeated using a lower level standard

(0.5 ppmC). If, after trying a second concentration level, the QC still does not meet acceptance

criteria, the task leader is contacted and the analyst and task leader discuss rerunning the

calibration curve.

After running and checking the QC, two zero air samples at 50% relative humidity are

injected to determine system cleanliness. The'average of the two area counts should be less than

10.0 ppbC. If the concentration is greater than 10.0 ppbC, an additional injection is analyzed and

the analyst averages all three. If the concentration is still greater than 10.0 ppbC, a leak check of

the analytical system is performed and the Task Leader is contacted.

glp/D:\SECT11.WPD

Project No. Element No. Revision No. Date

0121.00 B6

1 March 2000

3 of 11 Page

11.2 SNMOC Calibration

For the SNMOC instrument, a carbon response factor is obtained monthly based on the

analysis of humidified propane calibration standard. These standards are prepared by using the

Dynamic Flow Dilution System (SOP Number ERG-MOR-061) to dilute Scott Specialty or

Spectra Gas NIST certified standards into clean, evacuated stainless steel canisters. HPLC grade

water is injected to humidify the standard to approximately 75%. The standard is diluted with

nitrogen to achieve the desired concentrations for the calibration. The response factors generated

from the calibration are.used to determine concentrations in detected compounds, on the

assumption that FID response is linear with respect to the number of carbon atoms present in the

compound.

Calibration curve standards are made in ranges from 5 to 90 ppbC concentrations. The

calibration standards are analyzed in order of increasing concentration, and followed by the

system blank analysis to ensure no carryover after analysis of the high level standard. The

propane area count recorded by the FID is correlated to propane concentration by a least squares

linear regression and is used to quantitate the C2 through C,3 compounds. The calibration is

considered representative if the coefficient of correlation for the points from the curve and the

blank is greater than or equal to 0.995: for propane. The slopes of the regression lines are then

used to calculate monthly response factors.

Daily, before sample analysis, a QC standard, a certified PAMS standard prepared by

Scott Specialty or Spectra Gases is analyzed to ensure the validity of the current monthly

response factors. This standard has a midpoint concentration from the calibration for compounds

that span the carbon range. This level is considered representative of the majority of

concentrations expected in ambient air samples. The concentrations computed from the

QC standard are compared to the calculated theoretical concentrations. A concentration percent

bias of less than or equal to 30% is considered acceptable and the analytical system is in control.

glp/D:\SECT11.WPD

Project No. • 0121.00 Element No. B6 Revision No. 1 Date March 2000 Page 4 of 11

If the daily QC standard does not meet the 30% criterion, a second QC standard is

prepared and analyzed. If the second QC standard meets the criterion, the analytical system is

considered in control. If the second QC check does not pass, a leak test and system maintenance

is performed, and a third QC standard analysis is performed. I f the criterion is met by the third

analysis, the analytical system is considered in control. If the maintenance causes a change in

system response, a new calibration curve is required.

A system blank of cleaned, humidified air is analyzed after the daily QC standard

analysis and before sample analyses. The system is considered in control i f the total NMOC

concentration for the system blank is less than.or equal to 20 ppbC.

Retention time standards are used to gather information and set up a reference database

using relative retention times referenced to toluene. These relative retention times are used to

identify the target compounds in the ambient air samples.

For simplicity, each instrument is calibrated for all of the SNMOC, PAMS, and UATMP

compounds daily. All QC check standards have to pass each of the calibration procedures listed

in Sections 11.2, 11.3, and 11.4.

11.3 UATMP Calibration

Calibration of the GC/FID/MSD is accomplished by analyzing humidified calibration

standards generated from Scott Specialty or Spectra Gas certified standards. The certified

standards contain the UATMP target compounds at approximately 500 ppbv - polar compounds

at 1 ppm. Although the MSD is the primary quantitation tool, responses on the FID are recorded

and quantitated to detect and quantitate hydrocarbon peaks and can be used for SNMOC or

PAMS results. The calibration for these hydrocarbon peaks should be accomplished as explained

in Sections 11.2 and 11.4, respectively. >

glp/D:\SECT11.WPD

Project No. 0121.00 Element No. B6 Revision No. 1

; Date March 2000 Page 5 of 11

Calibration curves for the UATMP samples should include a five-point calibration and

daily calibration checks at a mid-level concentration for the target compounds (see Standard

Operating Procedure, ERG-MOR-061).

Calibration standards are generated with a dynamic flow dilution apparatus (Figure 11-1).

The gases are mixed in a SUMMA®-treated mixing sphere and bled into evacuated canisters.

One dilution air stream is routed through a SUMMA®- treated bubbler containing HPLC-grade

water to humidify; the other stream is not humidified. The dilution air streams are then brought

together for mixing with the streams from the certified cylinders. Flow rates from all streams are

gauged and controlled by mass flow controllers. The split air dilution streams are metered by

"wet" and "dry" rotameters from the humidified and unhumidified dilution air streams,-

respectively. Air is controlled from channel 4 where the mass flow controller ranges from

0-5 L/min, whereas all other channels range from 0-20 mL/min.

The system is evacuated with a vacuum pump while the closed canister is connected. The

lines leading to the canister and to the mixing sphere are flushed for at least 15 minutes with

standard gas before being connected to the canister for filling. A precision absolute pressure

gauge measures the canister pressure before and after filling.

Initial calibration curve standards are prepared at an average of 0.5, 1, 5, 10, and 15 ppbv

for each of the target compounds. All standards and samples are analyzed with the following

internal standards: «-hexane-d14, 1,4-difluorobenzene, and chlorobenzene-d5.

Bromofluorobenzene is also injected with the internal standards to verify mass spectrometer

tune. The calibration requires an average response factor, based on the internal standard, of

±30% relative standard deviation. The zero air used for canister cleaning and for standard

dilution is analyzed at the time of calibration, but the results are not included in the calibration

curve. Daily quality control verification is done with standards made from Scott certified gases

at an average concentration of 5 ppbv.

glp/D:\SECT11.WPD

CO m o

Absolute Pressure y ^ Gauge

Rotameter (exit)

Thermocouple

500-mL Summa® Treated Mixing Flask

All Tubing is Chromatographic Grade Stainless Steel All fittings are 316 Stainless Steel

Bellows Valve

Summa® Treated Canister

Mass Flow Controller

0:100 mUrn

Mass Flow Controller 0-20 mUm

Mass Flow Controller 0-20 mL/m

Mass Flow Controller 0-20 mL/m

Mass Flow Controller 0-20 mL/m

Mass Flow Controller 0-5 Urn

2.8-L Summa® Treated Humidifier

0,'s/g,'mO(,'33961099/15-1 .tif

O W tji <S ft ° ° « I 8

o 3 ~-

ON o

Figure 11-1. Dynamic Flow Dilution Apparatus

o cr O o o

o

to

W o Os O

Project No. Element No. Revision No. Date

0121.00 B6

1 March 2000

7 of 11 Page

For simplicity, each instrument is calibrated for all of the SNMOC, PAMS, and UATMP

compounds daily. All QC check standards have to pass each of the calibration procedures listed

in Sections 11.2, 11.3 and 11.4.

11.4 PAMS Calibration

The PAMS hydrocarbon analysis system is calibrated using the same procedure described

for SNMOC in Section 11.2 on the GC/FID/MSD system.

For the PAMS carbonyl analyses, the HPLC instrument is calibrated using 0.2 to

20 micrograms per milliliter (ug/mL) nominal concentrations of the derivatized targeted

compounds contained in a solution of acetonitrile. The: calibration curve consists of six

concentration levels between 0.2 and 20 ug/mL, and each is analyzed in replicate. The standard

linear regression analysis performed on the data for each analyte must have a correlation

coefficient greater than or equal to 0.995.

As a QC procedure to check HPLC column efficiency, a second source QC (SSQC)

sample solution containing 11 target carbonyl compounds at a known concentration is analyzed

after every calibration curve. A calibration accuracy check ( a midpoint calibration standard) is

analyzed after every 10 samples (meeting the ±15% criteria), and a system blank brackets the

analytical batch, by analyzing one blank at the beginning and one at the end.

11.5 HAPS Calibration

Analytical instruments and equipment are calibrated prior to each use or on a scheduled

periodic basis. Analytical methods requiring calibration standards are governed by Standard

Operating Procedures (SOPs) for laboratory standards.,; Calibration standards must be National

Institute of standards and Technology (NIST) traceable. Appropriate standards are prepared by

serial dilutions of pure substances or accurately prepared concentrated solutions. Many

glp/D:\SECT11.WPD

Project No. Element No. Revision No. Date

0121.00 B6

1 March 2000

8 of 11 Page

analytical instruments have high sensitivity, so calibration standards must be extremely dilute

solutions. In preparing stock solutions of calibration standards, great care is exercised in

measuring weights and volumes, since analyses following the calibration are based on the

accuracy of the calibration. Calibration requirements for the HAPS analytical methods are

shown in Table 11 -1.

glp/D:\SECT11.WPD

Project No. • 0121.00 Element No. • B6 Revision No. 1 Date March 2000 Page 9 of 11

Table 11-1

Analytical Equipment Calibration Requirements

Analytical Parameter

Quality Parameter Method of Determination

Frequency Acceptance Criteria

Particulate Matter Electronic Balance Calibrated against NIST Class S weights-

Post-test Within 2.0 mg

Metals - ICPMS ICAP Calibration-Quantitative

Initial analysis of 3 levels of standards bracketing sample concentrations

Twice per 3 runs Linear correlation coefficient >0.995

ICAP Calibration -Blanks

With calibration . standards

Every 10 samples Per manufacturer's specifications

ICAP Calibration -Interference Check

With calibration standards

Beginning and end of analyses

80-120% of expected value

ICAP Calibration -Continuing Check

Analysis of mid-range calibration standard

Once per 10 samples

90-110% of expected value

Ethylene Oxide -GC/ECD

GC/ECD Calibration -Quantitative

Analysis of 6 standards over the range of 1 to 42 u,g

Daily Nonlinear least squares fit in necessary to obtain the best fit.

GC/ECD Blanks With calibration standards

Every 10 samples Less than quantitative limit of 0.1 |ag

GC/ECD method spikes

Analysis of QC blind spikes

Per sample batch 80- 120% of expected value

glp/D:\SECT11.WPD

Project No. Element No. Revision No. Date Page

0121.00 B6

1 March 2000

lOof 11

Table 11-1

(Continued)

Analytical Quality Parameter Method of Frequency Acceptance Parameter Determination Criteria

PCDD/PCDF - Calibration - Initial analysis of Prior to sample Variability of mean HRGC/HRMS Quantitative standards at 5 levels analysis Response Factor

bracketing sample must be <25-20% concentrations Relative Standard

Deviation for each unlabeled analyte and internal standard and

; recovery standard. Signal/Noise ratio must be >2.5. Ion abundance ratios must be within control limits

Calibration - Analysis of column At start of each 12- Document Column Performance Check hr period resolution between Performance Check solution of 2,3,7,8-TCDD and

PCDD/PCDF other TCDDs (25% congeners valley)

Calibration - Analysis of mid- Every 12 hours Variability of mean continuing ' range calibration Response Factor

standard must be <25-20% Relative Standard Deviation for each unlabeled analyte and internal standard and

• recovery standards. Signal to Noise ratio must be > 2.5. Ion Abundance ratios must be within control limits.

glp/D:\SECT11.WPD

Project No. Element No. Revision No Date Page

Table 11-1

(Continued)

Analytical Parameter

Quality Parameter Method of Determination

Frequency Acceptance Criteria

Calibration Confirmation -Quantitative

Initial analysis of standards at 5 levels bracketing sample concentrations

Prior to sample analysis

Variability of mean Response Factor must be <25-20% Relative Standard Deviation for each unlabeled analyte and internal standard and recovery standards. Signal to Noise ratio must be > 2.5. Ion Abundance ratios must be within control limits.

Semivolatiles -GC/MS

Calibration -Quantitative

Initial analysis of standards at 5 levels bracketing sample concentrations

Prior to sample analysis

Variability of average Relative Response Factor < 30%

Calibration -Calibration Check Compounds

With calibration standards

Prior to sample analysis

Relative Response Factor MUST be <30%

Calibration -System Performance Check Compounds

With calibration standards

Prior to sample analysis

Minimum Relative Response Factor for Check Compounds > 0.050

Calibration - Daily calibration check

Every 12 hours Prior to sample analysis

Relative percent difference compared to mean of calibration curve <30%

Calibration - Blanks 20% of samples Concurrent with sample analysis

Analytes < Method Detection Limit

0121.00 B6

1 March 2000

11 of 11

Project No. 0121.00 Element No. B7 Revision No. 1 Date March 2000 Page 1 of 1

SECTION 12

DATA MANAGEMENT

All data generated in the ERG laboratory are collected on electronic tape or disk drives

and also paper copies. The printed copies of all reports are kept on file in the laboratory or in

storage. Final data are entered into Excel® and printed for the monthly or quarterly reports.

These reports are mailed to the EPA, State agencies, and participants. ERG will prepare a final

report containing all aspects of the program, including data summaries, QA, QC, and data

analysis results for the EPA, and distribute site-specific summaries of the final data to designated

State and local personnel.

glp/D:\SECT12.WPD

Project No. Element No. Revision No. Date

0121.00 CI

1 March 2000

1 of 3 Page

C—ASSESSMENT/OVERSIGHT

SECTION 13

ASSESSMENTS AND RESPONSE ACTIONS

13.1 QA Performance Audits

Quality assurance performance audit samples are provided by the EPA (or an EPA

contractor) as available. Percent accuracy (or bias) is calculated using the EPA-reported audit

sample concentration as the true value. For the NMOC program, audit samples of propane or

multicomponents in air are analyzed as received. Multi-component audit samples will be

analyzed for the 12-month UATMP by the GC/FID/MSD. For the SNMOC and PAMS

programs, multicomponent audit samples are also analyzed as received by the GC/FID/MSD

analytical system.

13.2 Performance Evaluation and System Audits

The Program Manager, Deputy Program Manager, Task Leader, and Program QA Officer

for ERG conduct performance and system audits on the laboratory procedures and the records

kept in the laboratory.

Program Manager, and Program Manager as needed. These reports are provided whenever a

QC problem that requires a change in the operating procedure occurs. These reports address

QC problems arising in the application of the work plan, an assessment of the probable

significance of the problems, and recommended corrective actions. QC problems to be addressed

may arise from:

13.3 QA Reports

The Program QA Officer provides written QA reports to the Task Leader, Deputy

glp/D:\SECT13.WPD

Project No. 0121.00 Element No. . CI Revision No. 1 Date March 2000 Page 2 of 3

• Poor compliance with sampling procedures reported by the project personnel;

• Invalid samples;

• In-process procedure changes required by the nature of the program; and

• Quality control waivers dictated by operating conditions.

The final report also addresses QA considerations of the whole project.

The assessment of the significance of the problems is based, in part, on the probable

effect on program completeness and validity of inferences made from the data.

Recommended actions include, as applicable:

• Tests that may clarify the problem;

• Corrective actions to alleviate the problem;

• Further documentation of the problem; and

• Acceptance of the anomalous condition with associated risk.

These reports will also include:

• Periodic assessment of measurement accuracy, precision, and completeness; and

• Results of performance and laboratory system audits.

In the final project report, a QA summary discusses all the QA activities and results for

the entire project.

glp/D':\SECT13.WPD

Project No. Element No. Revision No. Date

0121.00 C2

1 March 2000

1 of 2 Page

SECTION 14

REPORTS TO MANAGEMENT

14.1 QA and QC Functions

The lines of communication between management, the Program QA Officer, and the

technical staff are formally established and allow for discussion of real and potential problems,

preventive actions, and corrective procedures. The major QC responsibilities and QC review

functions are summarized in Table 14-1.

Anytime during the program, additional QA/QC measures may be initiated upon

consultation between the Task Leader, Program QA officer and the Senior Technical Advisor.

glp/D:\SECT14.WPD

Project No. 0121.00 Element No. C2 Revision No. 1 Date March 2000 Page 2 of2

Table 14-1

QC Responsibilities and Review Functions

Responsible Person Major Responsibilities

Program Manager • Ensure overall timely performance of high quality technical services • Communicate technical issues and needs • Track all management systems and tools • Track deliverables and budget performance • Review reports before reporting to the client

Deputy Program Manager

• Ensure data quality • Check information completeness • Assist with technical problems • Ensure appropriate level of staffing and committed resources exist to perform

work • Review data completeness and quality before reporting to client

Review all reports • Report project performance (budget and deliverables) to EPA at monthly

meetings and in monthly progress reports • ' Day-to-day management of task leaders

Program QA Officer Review QC reports • Make QA recommendations • Write and/or review test plan • Write and/or review QAPP • Audit laboratory(s) • Review documentation (reports, etc.)

Technical Advisor • Propose procedural change Propose equipment change

• Assist with technical problems

Peer Reviewer • Ensure final data quality • Final data review ' • Assist with technical problems

Analytical Task Leader • Review documentation • Develop analytical procedures

Propose procedural changes • Data review and validation • . Analyst training and supervision

Meet-task budgets and report schedules • Manage day-to-day technical activities • Check information completeness

Review instrument and maintenance log books • Review calibration factor drift • Perform preventive maintenance • Prepare monthly/quarterly reports

glp/D:\SECT14.WPD

Project No. Element No. Revision No. Date

0121.00 Dl • 1

March 2000 1 of 9 Page

D—DATA VALIDATION AND USABILITY

SECTION 15

DATA REVIEW, VALIDATION, AND VERIFICATION REQUIREMENTS

ERG uses several software programs maintained on microcomputers for data storage,

retrieval, analysis, and reporting. These programs include Microsoft Excel®; Access®; and SAS®.

Data summaries, QC charts, and other graphs, generated in a cost-effective manner, aid in

maintaining consistent data quality.

Each sample received at the ERG Laboratory is logged into the sample logbook and into

the computerized login. The accompanying field data forms are reviewed to verify that all forms

are complete.

The reliability and acceptability of environmental analytical information depends on the

rigorous completion of all the requirements outlined in. the QA/QC protocol. During data

analysis and validation, data are filtered and accepted or rejected based on the set of QC criteria.

The data are critically reviewed to locate and isolate spurious values. This review may involve

only a cursory scan to detect extreme values or a detailed evaluation requiring the use of a

computer. In either case, when a spurious value is located, it is not immediately rejected. All

questionable data, whether rejected or not, are maintained along with rejection criteria and any

possible explanation. A detailed approach such as this can be time-consuming but can also be

helpful in identifying sources of error and, in. the long run, save time by reducing the number of

outliers.

Prior to any statistical approach, the reported data are checked to ensure accurate

transcription. The value is double-checked and a comparison to previously recorded data is

made. Using conveniently formatted and bound prepared data recording forms is essential;

hardcopies of data can also be obtained directly from measuring devices that are equipped with

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 Dl

1 March 2000

2 of 9 Page

the necessary digital recording peripheral. Usually, this method of recording data is sufficient if

the hardcopies are properly labeled and filed, although periodic checks should be performed to

ensure the proper operation of such a device.

The collected data are reviewed by the analyst and the task leader. The data are

scrutinized daily to eliminate the collection of invalid data. The analyst records any unusual

instances (no matter how minor) during analysis (e.g., power loss or fluctuations, temporaiy

leaks or adjustments, operator error) on the chain of custody forms and. notifies the analytical

task lead.

15.1 NMOC/SNMOC Data Reduction, Validation, and Reporting

Analytical data forms have been developed for samples receiving analysis. A copy of the

field data sheet is attached to the analytical form. The analytical data form includes site

collection information from the field data sheet, as well as analysis information and results; this

information is transferred to computer programs, including Excel®, for data storage, retrieval,

analysis,, and reporting. The analytical data forms (with attached field data sheet) are stored in

notebooks or folders in order of ID number.

15.1.1 NMOC Data Reduction. Validation, and Reporting

Monthly site-specific NMOC data update summaries are developed for the purpose of

distribution to the participating EPA technical,staff, administrators, and to the administrators of

the State agencies involved in the study. Each summary updates prior data listings. Cumulative

listings are periodically generated upon request. Even though these data summaries have not

passed through the final data validation steps, this timely turnaround of NMOC data assists in

planning, preliminary modeling, and program development for the participating State agencies.

Any changes made in the preliminary data as a result of subsequent data validation processes are

noted in the cumulative project data summaries for each specific sampling site.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date March 2000

3 of 9

0121.00 Dl

Page

Ten percent of the NMOC database is checked to verify its validity. Items checked

include original data sheets, checks of all calculations (from calibration to sample analysis), and

data transfers. Corrections are made to the database as errors or omissions are encountered. The

analytical reviewer examines all data for overall data quality and completeness. The Deputy

Program Manager reviews all data before data are reported to the EPA or the states.

A final report containing all aspects of the NMOC program including data summaries,

QA results, QC results, and data analysis results is prepared for EPA. Site-specific data

summaries are prepared and distributed to designated State and local agencies. The final

NMOC data are submitted to the Aerometric Information Retrieval System (AIRS) Air Quality

Subsystem (AQS) as detailed in Section 15:4

15.1.2 SNMOC Data Reduction. Validation, and Reporting

A sample analysis logbook is maintained to detail pertinent sample information at the

time of analysis. Entries include site code, sample date, analysis date, and electronic file names.

A PE Nelson Turbochrom® Navigator Data System consisting of a 900 Series Intelligent

Interface and a PC system containing the Turbochrom® software is used to acquire, integrate, and

store the analytical data. A chromatograph and area count report from each detector are printed

for each analysis. Electronic copies of the data are stored on 1.44 MB flexible disk cartridges

and a compressed backup tape (250MB).

The data are processed using PE Nelson Turbochrom® version 4.1 software. The

software uses a database containing relative retention time information for all compounds of

interest and applicable response factors to process the data files. A preliminary report is

generated containing possible peak identifications and quantitations based on the carbon

response factors for propane in effect at the time of analysis.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 Dl

1 March 2000

4 of 9 Page

A data reviewer compares the Turbochrom raw data report to the chromatogram to

determine proper peak identifications. A second data review is performed to check for items that

may have been overlooked on the first pass. After the data are reviewed twice, a final report, in

Excel® format, is processed and reviewed for completeness. Final report versions containing

information on all quantitated peaks are printed and filed with the analysis chromatogram

printout and preliminary Excel® report. Electronic copies of all Excel® reports are kept on file.

The analytical reviewer examines all data for overall data quality and completeness. The Deputy

Program Manager reviews all data before it is reported to the EPA or the states. These

procedures are outlined in the laboratory SOP (ERG-MOR-005) for Sample Analysis and

Validation of Hydrocarbons on the UATMP system with a GC/FID detector.

A final report containing all aspects of the SNMOC program (including data summaries,

QA, QC, and data analysis results) is prepared for EPA. Site-specific data summaries are

prepared and distributed to designated State and local contacts. The final SNMOC data are

submitted to the AIRS AQS.

15.2 UATMP Data Reduction, Validation, and Reporting

A sample analysis logbook is maintained to detail pertinent sample information at the

time of analysis. Entries include site code, sample data, analysis date, and electronic file names.

A Hewlett Packard Chemstation® and PE Turbochrom® Data System consisting of a

900 Series Intelligent Interface and a PC system containing the software are used to acquire,

integrate, and store the analytical data. The MSD data are reported with the chromatograph and

detailed information. A chromatograph and area count report from each detector are printed for

each analysis. Electronic copies of the data are stored on tape or disk.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 Dl

1 March 2000

5 of 9 Page

The task leader reviews all of the generated analytical reports to verify peak

identifications. Printed copies of all reports are kept on file in the laboratory or in storage. Final

data are entered into Excel® and printed for the quarterly or monthly reports.

Quarterly GC/FID/MSD data summaries are developed for distribution to the

participating EPA technical staff, administrators, and administrators of the State agencies

involved in the study. The data summaries include:

'• Site code

• • Sample identifications >

• Sample dates

• Target compound list

• Concentrations (ppbv)

Quarterly preliminary data summaries are mailed to the State agencies and participants. These

data summaries are considered preliminary until the final report, at which time the data are

validated.

The analytical reviewer examines all data for overall data quality and completeness. The

Deputy Program Manager reviews all data before they are reported to the EPA or the states.

ERG prepares a final report containing all aspects of the UATMP including data summaries, QA,

QC, and data analysis results for EPA, and distribute site-specific summaries of the final data to

designated State and local personnel. ERG staff follow the SOP for the Concurrent

GC/FID/MSD Analysis of Canister Air Toxic Samples (ERG-MOR-005). ERG will submit the

final UATMP data to the AIRS AQS, as detailed in Section 15.4.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 Dl

1 March 2000

6 of 9 Page

15.3 PAMS Data Reduction, Validation, and Reporting

15.3.1 VOC Data

Data from PAMS VOC hydrocarbon analyses performed at ERG are processed using the

same procedures described for UATMP in Section 15.2. The final data are submitted to the

AIRS AQS as detailed in Section 15.4. For the PAMS sites, there is an option of statistically

validating the data generated using software generated by using Sonoma Technical Institute'

(VOCDat).

15.3.2 Carbonyl Compounds Data

All carbonyl samples received are given an ID number that corresponds to the

VOC canister sample. An extraction log is maintained to record pertinent information at the time

of extraction. A sample analysis log is also maintained to record peilinent information at the

time of analysis.

A PE Turbochrom® Data System is used to acquire, integrate, quantitate, and store the

analytical data. Preliminary peak identifications are determined based on elution times. A data

reviewer compares the chromatogram and the QC chromatogram to determine proper peak

identifications and determine i f reintegration is needed on any peak. Quantitations are based on

raw amounts of analyte in pg/mL calculated by the Turbochrom Data System from a 6-point,

least-squares regression. Results in ppbv are then calculated as described in Method TO-11A.

The analytical reviewer examines all data for overall data quality and completeness. The Deputy

Program Manager reviews all data before they are reported to the EPA or the states. Final report

versions containing information on all quantitated peaks are printed using spreadsheet software,

and the final data is submitted to the AIRS AQS as detailed in Section 15.4.

15.4 HAPS Data Reduction, Validation, and Reporting

glp/D:\SECT15.WPD

Project No. 0121.00 Element No. Dl Revision No. 1 Date March 2000 Page 7 of9

The HAPS analytical procedures performed during the monitoring program will be

checked against those described in the QAPP. Deviations from the QAPP will be classified as

acceptable or unacceptable, and critical or noncritical. Acceptance criteria are stated in each

method and in Section 8 of this document. The critical or noncritical nature of a deviation will

be determined in the DQA process.

Quality control samples and procedures performed during the monitoring program will be

checked against those described in Section 4 of the QAPP. Omissions will be discussed in the

final report. Quality control results (matrix/method spike recoveries, blank analysis, duplicate

analysis, etc.) will be reviewed. All results outside specified parameters will be discussed with

the EPA Delivery Order Manager for corrective action. In some cases, reference methods have

guidance on corrective action. Where available, the guidance in the reference methods will be

followed.

Documentation of equipment and instrument calibration (e.g., monitoring equipment and

analytical instalments) will be checked against' the values used in data collection. Errors and

omissions will be discussed in the final report. The documentation will be checked to ensure that

the calibration: •

• Was performed within an acceptable time prior to the sampling dates;

• Includes the proper number of calibration points;

• Was performed using appropriate standards for the reported measurements;

• Had acceptable checks to ensure that the measurement system or analytical system was stable when the calibration was performed.

The data processing systems will be checked by using raw data for which calculated

values are already known. The example data will be put into the system and the calculated

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date March 2000

8 of 9

0121.00 Dl

Page

results compared to the known values. Hand calculations will be used to check the data

processing system. Findings from these audits will be included in the final report.

15.5 Aerometric Information Retrieval System Air Quality Subsystem (AIRS AQS)

ERG submits all data collected for the NMOC, UATMP, and PAMS programs to the

AIRS AQS Subsystem.

Prior to ERG's submittal of data to AIRS, the State or local agency submits monitor

transactions (Type Fl) or ensures that monitor transactions (Type Fl) have been submitted.

ERG supplies the State or local agency with assistance concerning parameter coding for this

submittal. The Type Fl cards prepare the AIRS AQS for the raw data transactions (Type 1 or 2).

The AIRS submittal process involves the following six steps:

The raw data are formatted to comply with the requirements of AIRS AQS. The hourly data (sampling intervals of less than 24 hours) are formatted to comply with the hourly file (Type 1). The daily data (sampling intervals of 24 hours) are formatted to comply with the daily file (Type 2).

The Type 1 transaction files and Type 2 transaction files are reviewed to ensure that proper monitor ID (including state, county, site, parameter, and POC codes), interval, units, method, date, start hour, decimal point indicators, and sample value codes are correct.

The transaction files are loaded into the screening,file currently assigned to ERG.

The transaction files in the screening file are edited, the records examined in the screening file, and the data validated. The three edit checks ensure that proper codes have been used, that proper relationships exist (e.g., the reported compound and method agree), and that proper relationships exist with the data currently in the AIRS database.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 Dl

1 March 2000

9 of 9 Page

The transaction files in the screening file are corrected. Any records that do not pass the edit checks are corrected. Edit checks are then performed on the corrected data.

The AIRS Database Administrator is notified that the transactions in the screening are ready to be used in an update.

glp/D:\SECT15.WPD

Project No. Element No. Revision No. Date

0121.00 D2

1 March 2000

1 of 1 Page

SECTION 16

RECONCILIATION WITH DATA QUALITY OBJECTIVES

The project management team, quality assurance officer, and sampling and analytical

team members are responsible for ensuring that all measurement procedures are followed as

specified and that measurement data meet the prescribed acceptance criteria. Prompt action is

taken to correct any problem that may arise.

QC problems requiring major corrective action are documented. The Program

QA Officer or other project members initiate corrective action if QC results exceed control

limits, or i f another problem or potential problem is identified. Corrective action is immediately

reported in a corrective action report to appropriate project management and the Program

QA Officer. Corrective action is also initiated by the Program QA Officer based on QC data or

audit results.

' In addition to the corrective action reporting system for addressing problems identified

through the internal QC system, a system for issuing recommendations for corrective action

exists for addressing problems identified through QA review. Each recommendation addresses a

specific problem or deficiency.

Each of these written recommendations requires a written response from the responsible

party. Each also requires the Program QA Officer to respond and verify that the corrective action

has been implemented.

glp/D:\SECT16.WPD

Project No. 0121.00 Revision No. 1 Date March 2000 Page • 1 of3

SECTION 17

REFERENCES

1. McElroy, F.F., V.L. Thompson, and H.G. Richter. A Cryogenic Preconcentration-Direct FID (PDFID) Method for Measurement of NMOC in Ambient Air, EPA-600/4-85-063. Research Triangle Park, NC: U.S. Environmental Protection Agency, 1985.

2. McAllister, R. A., D-P. Dayton, and D. E. Wagoner. 1985 Nonmethane Organic Compounds Monitoring Assistance for Certain States in EPA Regions I , III , V, VI, and VII. Radian Corporation, DCN No. 85-203-024-35-01, prepared for Dr. Harold G. Richter, Research Triangle Park, NC: U. S. Environmental Protection Agency, 1986.

3. Purdue, L.J., D-P. Dayton, and J. Rice. Technical Assistance Document for Sampling and Analysis of Ozone Precursors. EPA 600/8-91-215. Research Triangle Park, NC: U.S. Environmental Protection Agency, 1991. Revised in 1994 and 1998.

4. McAllister, R. A., D-P. Dayton, and D. E. Wagoner. Nonmethane Organic Compounds Monitoring Assistance for Certain States in EPA Regions III. IV, V, VI, and VII. Phase I I . Radian Corporation, DCN No. 85-203-024-12-02, prepared for Dr. Harold G. Richter, Research Triangle Park, NC: U.S. Environmental Protection Agency, 1985.

5. McAllister, R. A., R. F. Jongleux, D-P. Dayton, P. L. O'Hara, and D. E. Wagoner. 1986 Nonmethane Organic Compounds Monitoring. Radian Corporation, DCN No. 87-203-024-93-11, prepared for Dr. Harold G. Richter, Research Triangle Park, NC: U. S. Environmental Protection Agency, 1987.

6. McAllister, R.A., P.L. O'Hara, D.E. Wagoner, D-P. Dayton, R.F. Jongleux. 1987 Nonmethane Organic Compound and Air Toxics Monitoring Program. Volumes I and II . Radian Corporation, DCN No. 87-203-024-93-11, prepared for Dr. Harold G. Richter, Research Triangle Park, NC: U. S. Environmental Protection Agency, 1988.

7. McAllister, R.A., W.H. Moore, D-P. Dayton, J. Rice, R.F. Jongleux, R.G. Merrill, Jr., J.T. Bursey, and P.L. O'Hara. 1989 Nonmethane Organic Compound and Urban Air Toxics Monitoring Programs. Final Report. Volume I . Nonmethane Organic Compound and Three-Hour Air Toxics Monitoring Programs. Radian Coiporation, DCN No. 88-262-045-25, prepared for Dr. Harold G. Richter, Research Triangle Park, NC: U. S. Environmental Protection Agency, 1988.

glp/D:\SECT17.WPD

Project No. Revision No. Date Page

0121.00 1

March 2000 2 of 3

9.

10.

11.

12.

13.

14.

McAllister, R.A., B.W. Nelson, W.H. Moore, D-P. Dayton, .1. Rice, R.F. Jongleux, R.G. Merrill, Jr., J.T. Bursey, and P.L. O'Hara. 1989 Nonmethane Organic Compound and Three-Hour Air Toxics Monitoring Program, Final Report. Radian Corporation, DCN No. 262-045-89, prepared for Neil J. Berg, Jr., Research Triangle Park, NC: U. S. Environmental Protection Agency, 1990.

O'Hara, P.L., R.A. McAllister, D-P. Dayton, J.E. Robbins, R.F. Jongleux, R.G. Merrill, Jr., J. Rice, J.E. McCartney, T.L. Sampson, and J.Y. Martin. 7997 Nonmethane Organic Compound, Speciated Nonmethane Organic Compound, and Three-Hour Air Toxics Monitoring Program, Final Report. Radian Corporation, DCN No. 92-262-045-90, prepared for Neil J. Berg, Jr., Research Triangle Park, NC: U. S. Environmental Protection Agency, 1992.

O'Hara, P.L.,.R.G. Merrill, Jr., T.L. Sampson, D-P. Dayton, J. Rice, J.E. McCartney, and J.Y. Martin. 1992 Nonmethane Organic Compounds and Speciated Nonmethane Organic Compounds Monitoring Programs, Final Report. Radian Corporation, DCN No. 93-298-017-70-13, prepared for Neil J. Berg, Jr., Research Triangle Park, NC: U. S. Environmental Protection Agency, 1993.

Dayton, D-P., R. A. McAllister, D. Wagoner, F. F. McElroy, V. L. Thompson, and H. G. Richter, U. S. Environmental Protection Agency, "An Air Sampling System for Measurement of Ambient Organic Compounds," Paper presented at the 1986 U.S. EPA/APCA Symposium: Measurement of Toxic Air Pollutants, Raleigh, NC, April 27-30, 1986.

1993 Nonmethane Organic Compounds and Speciated Nonmethane Organic Compounds Monitoring Programs, Final Report. Radian Corporation, DCN No. 93-298-130-12-10, prepared for Neil J. Berg, Jr., Research,Triangle Park, NC: U. S. Environmental Protection Agency, 1994.

1994 Urban Air Toxics Monitoring Program, Final Report. Radian Corporation. Prepared for Kathy Weant and Neil J. Berg, Jr., Research Triangle Park, NC: U..S. Environmental Protection Agency, 1996.

Steger, J., and J. Rice. 1995 Non-Methane Organic Compounds and Speciated Non-Methane Organic Compounds Monitoring Programs, Final Report. Eastern Research Group, Inc., prepared for Kathy Weant and Neil J. Berg, Jr., Research Triangle Park, NC: U. S. Environmental Protection Agency, 1996.

1995 Urban Air Toxics Monitoring Program, Final Report. Eastern Research Group, Inc., prepared for Kathy Weant and Neil J. Berg, Jr., Research Triangle Park, NC: U. S. Environmental Protection Agency, 1997.

glp/D:\SECT17.WPD

APPENDIX C

U.S. Environmental Protection Agency (EPA)

Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specialty-Prepared

Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS) from the

Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air Second Edition

Center for Environmental Research Information

Office of Research and Development. Cincinnati, OH January 1999 :

U'.-VOiD/K-vO/OlOb

Compendium of Methods for the Determination of

Toxic Organic Compounds in Ambient Air

i

l

Second Edition ; ! i i i i

Compendium Method TO-15

Determination Of Volatile Organic Compounds (VOCs) In Air Collected In

Specially-Prepared Canisters And Analyzed By Gas Chromatography/

Mass Spectrometry (GC/MS)

Center for Environmental Research Information Office of Research and Development

U.S. Environmental, Protection Agency Cincinnati, OH 45268

January 1999

METHOD TO-15

Determination of Volatile Organic Compounds (yOCs) In Air Collected In Specially-Prepared Canisters And Analyzed By Gas Chromatography/

Mass Spectrometry (GC/MS)

TABLE OF CONTENTS

Page

1. Scope

2. Summary of Method 1>2

. 3. Significance \ ' • 15-3

4. Applicable Documents 4.1 ASTM Standards ' . ' . . ' ' [[ ' [ [ [ " ' ] ' ' [ ' ' " " .' 4.2 EPA Documents , ~

15-4 5. Definitions

' ''. ; lD-4 6. Interferences and Contamination ... , c r

' ' '• I 5-6 7. Apparatus and Reagents ^ ^

7.1 Sampling Apparatus : J 5 6

7.2 Analytical Apparatus •. . j 5 §

7.3 Calibration System and Manifold Apparatus 15 10 7.4 Reagents..... \\\\\\\\\\\\\\[ [ [ [ [ [ [ [ ';' ^

8. - . Collection of Samples in Canisters 15 10 8.1 Introduction 15 10 8.2 Sampling System Description 1 5 1 1

8.3 Sampling Procedure 15 12 8.4 Cleaning and Certification Program 1' ' 15 14

9. GC/MS Analysis of Volatiles from Canisters j 5_, 6

9.1 Introduction . 15 16 9.2 Preparation of Standards ) 5 " ] 7

10. GC/MS Operating Conditions 1 5 9 ]

10.1 Preconcentrator . , " 10.2 GC/MS System I'"99 10.3 Analytical Sequence • [ 5

10.4 Instrument Performance Check . 1 5 ? ~ 10.5 Initial Calibration \S~2n

10.6 Daily Calibration 1 5 9 ^ 10.7 Blank Analyses : ) 5 9 7

10.8 Sample Analysis >cZ0

It!

J C . in

METHOD TO-15

Determination of Volatile Organic Compounds (VOCs) In Air Collected In Specially-Prepared Canisters And Analyzed By Gas Chromatography/

Mass Spectrometry (GC/MS)

1. Scope

1.1 Tins method documents sampling and analytical procedures for the measurement of subsets of the 97 volatile organic compounds (VOCs) that are included in the 189 hazardous air pollutants (HAPs) listed in Title HI of the Clean Air Act Amendments of 1990. VOCs are defined hereas organic compounds having a vapor press

• greater than 10;' Torr at 25'C and 760 mm Hg. Table 1 is the list of the target VOCs alone wit^heTcAS number, boiling point, vapor pressure and an indication of their membership in both the list of VOCs covered by Compendium Method TO-14A (I) and the list of .VOCs in EPA's Contract Laboratory Program (CLP) document entitled: Statement-of-Work (SOW) for the Analysis of Air Toxics from Superfund Sites (2).

Many of these compounds have been tested for stability in concentration when stored in specially-prepared canisters (see Section 8) under conditions typical of those encountered in routine ambient air analysis The stability of these compounds under all possible conditions is not known. However, a model to predict compound los.es due to physical adsorption of VOCs on canister walls and to dissolution of VOCs in water condensed in the canisters has been developed (3). Losses due to physical adsorption require oniv the S s h m e m of equihonum between the condensed and gas phases and are generally considered short term losses (i occurring over minutes to hours). Losses due to chemical reactions of the VOCs with collected ozone or other gas phase species also account for some short term losses. Chemical reactions between VOCs and substances ins.de the canister are generally assumed to cause the gradual decrease of concentration over time (i e Ion* term losses over days to weeks). Loss mechanisms such as aqueous hydrolysis and biological degradation (4) also exist. No models are currently known to be available to estimate and characterize all these potential losses although a number of experimental observations are referenced in Section 8. Some of the VOCs listed in Title III have short atmospheric lifetimes and may not be present except near sources.

1.2 This method applies to ambient concentrations of VOCs above 0.5 ppbv and tvpically requires VOC enrichment by concentrating up to one liter of a sample volume. The VOC concentration range for ambient air in many cases includes the concentration-at which continuous exposure over a lifetime is estimated to constitute a 10 or higher lifetime>nsk of developing cancer in humans. Under circumstances in which many hazardous VOCs are present at 10"6 risk concentrations, the total risk may be significantly greater.

13 This method applies under most conditions encountered in sampling of ambient air into canisters However the composition ot a gas mixture in a canister, under unique or unusual conditions, will change so that the sample is known not to be a true representation of the ambient air from which it was taken. For example, low humiditv condi ons in the sample may ead to losses of certain VOCs on the canister walls, losses that wo Id not happen if the humidity were higher: If the canister is pressurized, then condensation of water from high humidity sample may cause fractional losses of water-soluble compounds: Since the canister surface area is limited aliases are ..i competition for the available active sites. Hence an absolute storage stability cannot be assigned to a%ecitic gas. Fortunately, under conditions of normal usage for sampling ambient air. most VOCs can be recovered from canisters near their original concentrations after storage times of up to thirty days (see'Section 8).

1.4 Use of the Compendium Method TO-15 for many of the VOCs listed in Table 1 is likely to present two difficulties: (I) what calibration standard to use for establishing a basis for testing and quantitation, and (2) how

January 1999 Compendium of Methods tor Toxic Organic Air Pollutants P a g e 15-1

VOCs Method TO-15

reducing the sample size. For example, a small sample can be concentrated on a cold trap and released directly to the gas chromatographic column. The reduction in sample volume may require an enhancement of detector sensitivity.

Other water management approaches are also acceptable as long as their use does not compromise the attainment of the .performance criteria listed in Section II." A listing of some commercial water management systems is provided in Appendix A. One of the alternative ways to dry the sample is to separate VOCs from condensate on a low temperature trap by heating and purging the trap.

2.5 The analytical strategy for Compendium Method TO-15 involves using a high resolution gas chromatograph (GC) coupled to a mass spectrometer. If the mass spectrometer is a linear quadrupole system, ft is operated either by continuously scanning a wide range of mass to charge ratios (SCAN mode) or by monitoring select ion monitoring mode (SIM) of compounds on the target list. If the mass spectrometer is based on a standard ion trap design, only a scanning mode is used (note however, that the Selected Ion Storage (SIS) mode for the ion trap has features of the SIM mode). Mass spectra for individual peaks in the total ion chromatogram are examined with respect to the fragmentation pattern of ions corresponding to various VOCs including the intensity of primary and secondary ions. The fragmentation pattern is compared with stored spectra taken under similar conditions, in order to identify the compound. For any given compound, the intensity of the primary fragment is compared with the system response to the primary fragment for known amounts of the compound. This establishes the compound concentration that exists in the sample. . ' ,

Mass spectrometry is considered a more definitive identification technique than single specific detectors such as flame ionization detector (FID), electron capture detector (ECD), photoionization detector (PID), or a multidetector arrangement of these (see discussion in Compendium Method TO-14A). The use of both gas chromatographic retention time and the generally unique mass fragmentation patterns reduce the chances for misidentification. If the technique is supported by a comprehensive mass spectral database and a knowledgeable operator, then the correct identification and quantification of VOCs is further enhanced.

3. Significance

3.1 Compendium Method TO-15 is significant in that it extends the Compendium Method TO- 14A description for using canister-based sampling and gas chromatographic analysis in the following ways:

• Compendium Method TO-15 incorporates a multisorbent/dry purge technique or equivalent (see Appendix A) for water management thereby addressing a more extensive set of compounds (the VOCs mentioned in Title III of the CAAA of 1990) than addressed by Compendium Method TO-14A. Compendium Method TO-14A approach to water management alters the structure or reduces the sample stream concentration of some VOCs, especially water-soluble VOCs.

• Compendium Method TO-15 uses the GC/MS technique as the only means to identify and quantitate target compounds/The GC/MS approach provides a more scientifically-defensible detection scheme which is generally more desirable than the use of single or even multiple specific detectors.

• In addition, Compendium Method TO-15 establishes method performance criteria for acceptance of data, allowing the use of alternate but equivalent sampling and analytical equipment. There are several new and viable commercial approaches for water management as noted in Appendix A of this method on which to base a VOC monitoring technique as well as other approaches to sampling (i.e., autoGCs and solid

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-3

5.2 Absolute Pressure—pressure measured with reference to absolute zero pressure, usually expressed in units ofkPa,orpsi.

5.3 Cryogen—a refrigerant used to obtain sub-ambient temperatures in the VOC concentrator and/or on front of the analytical column. Typical cryogens are liquid nitrogen (bp -195.8°C), liquid argon (bp -185 7°C) and liquid CO: (bp-79.5°C).

5.4 Dynamic Calibration—calibration of an analytical system using calibration gas standard concentrations in a form identical or very similar to the samples to be analyzed and by introducing such standards into the inlet of the sampling or analytical system from a manifold through which the gas standards are flowing.

5.5 Dynamic Dilution—means of preparing calibration mixtures in which standard gas(es) from pressurized cylinders are continuously blended with humidified zero air in a manifold so that a flowing stream of calibration mixture is available at the inlet of the analytical system.

5.6 MS-SCAN—mass spectrometric mode of operation in which the gas chromatograph (GC) is coupled to a mass spectrometer (MS) programmed to SCAN all ions repeatedly over a specified mass range.

5.7 MS-SIM—mass spectrometric mode of operation in which the GC is coupled to a MS that is programmed to scan a selected number of ions repeatedly [i.e., selected ion monitoring (SIM) mode].

5.8 Qualitative Accuracy—the degree of measurement accuracy required to correctly identify compounds with an analytical system.

5.9 Quantitative Accuracy—the degree of measurement accuracy required to correctly measure the concentration of an identified compound with an analytical system with known uncertainty.

5.10 Replicate Precision—precision determined from two canisters filled from the same air mass over the same time period and determined as the absolute value of the difference between the analyses of canisters divided by their average value and expressed as a percentage (see Section 11 for performance criteria for replicate precision).

5.11 Duplicate Precision—precision determined from the analysis of two samples taken from the same canister. The duplicate precision is determined as the absolute value of the difference between the canister analyses divided by their average value and expressed as a percentage.

5.12 Audit Accuracy—the difference between the analysis of a sample provided in an audit canister and the nominal value as determined by the audit authority, divided by the audit value and expressed as a percentage (see Section 11 for performance criteria for audit accuracy).

6. Interferences and Contamination

6.1 Very.volatile compounds, such as chloromethane and vinyl chloride can display peak broadening and co-elution with other species if the compounds are not delivered to the GC column in a small volume of carrier gas. Refocusing of the sample after collection on the primary trap, either on a separate focusing trap or at the head of the gas chromatographic column, mitigates this problem.

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-5

VOCs Method TO-15

7.1.1.6. Electronic Timer. For unattended sample collection. 7.1.1.7 Solenoid Valve. Electrically-operated, bi-stable solenoid valve with Viton® seat and O-rings. A

Skinner Magnelatch valve is used for purposes of illustration in the text (see Figure 2). 7.1.1.8 Chromatographic Grade Stainless Steel Tubing and Fittings. For interconnections. All such

materials in contact with sample, analyte, and support gases prior to analysis should be chromatographic grade stainless steel or equivalent. "

7.1.1.9 Thermostatically Controlled Heater. To maintain above ambient temperature inside insulated sampler enclosure.

. 7.1.1.10 Heater Thermostat. Automatically regulates heater temperature. 7.1.1.11 Fan. For cooling sampling system. 7.1.1.12 Fan Thermostat. Automatically regulates fan operation. 7.1.1.13 Maximum-Minimum Thermometer. Records highest and lowest temperatures during sampling

period. 7.1.1.14 Stainless Steel Shut-off Valve. Leak free, for vacuum/pressure gauge. 7.1.1.15 Auxiliary Vacuum Pump. Continuously draws air through the inlet manifold at 10 L/min. or

higher flow rate. Sample is extracted from the manifold at a lower rate, and excess air is exhausted.

[Note: The use of higher inlet flow rates dilutes any contamination present in the inlet and reduces the possibility of sample contamination as a result of contact with active adsorption sites on inlet walls.]

7.1.1.16 Elapsed Time Meter. Measures duration of sampling. 7.1.1.17 Optional Fixed Orifice, Capillary, or Adjustable Micrometering Valve. May be used in lieu

of the electronic flow controller for grab samples or short duration time-integrated samples. Usually appropriate only in situations where screening samples are taken to assess future sampling activity.

7.1.2 Pressurized (see Figure 1 with metal bellows type pump and Figure 3). 7.1.2.1 Sample Pump. Stainless steel, metal bellows type, capable of 2 atmospheres output pressure.

Pump must be free of leaks, clean, and uncontaminated by oil or organic compounds.

[Note: An alternative sampling system has been developed by Dr. R. Rasmussen, The Oregon Graduate Institute of Science and Technology, 20000 N.W. Walker Rd., Beaverton, Oregon 97006, 503-690-1077, and is illustrated in Figure 3. This flow system uses, in order, a pump, a mechanical flow regulator, and a mechanical compensation flow restrictive device. In this configuration the pump is purged with a large sample flow, thereby eliminating the need for an auxiliary vacuum pump to flush the sample inlet.]

7.1.2.2 Other Supporting Materials. All other components of the pressurized sampling system are similar to components discussed in Sections 7.1.1.1 through 7.1.1.17.

7.2 Analytical Apparatus

7.2.1 Sampling/Concentrator System (many commercial alternatives are available). 7.2.1.1 Electronic Mass Flow Controllers. Used to maintain constant-flow (for purge gas, carrier gas

and sample gas) and to provide an analog output to monitor flow anomalies. 7.2.1.2 Vacuum Pump. General purpose laboratory pump, capable of reducing the downstream pressure

of the flow controller to provide the pressure differential necessary to maintain controlled flow rates of sample air. !

7.2.1.3 Stainless Steel Tubing and Stainless Steel Fittings. Coated with fused silica to minimize active adsorption sites. • *

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-7

VOCs Method TO-15

parallel array of rods under the influence of the generated field. Ions which are successfully transmitted through the quadrupole are said to possess stable trajectories and are subsequently recorded with the detection system. When the DC potential is zero, a wide band ofm/z values is' transmitted through the quadrupole. This "RF only" mode is referred to as the "total-ion" mode. In this mode, the quadrupole acts as a strong focusing lens analogous to a high pass filter. The amplitude of the RF determines the low mass cutoff. A mass~spectrum is generated by scanning the DC and RF voltages using a fixed DC/RF ratio and a.constant drive frequency or bv scanning the frequency and holding the DC and RF constant.. With the quadrupole system only 0.1 to 0.2 percent ofthelons formed in the ion source actually reach the detector.

7.2.2.3.2Ion Trap Technology. An ion-trap mass spectrometer consists of a chamber formed between two metal surfaces in the shape of a hyperboloid of one sheet (ring electrode) and a hyperboloid of two sheets (the two end-cap electrodes). Ions are created within the chamber by electron impact from an electron beam admitted through a small aperture in one of the end caps.' Radio frequency (RF) (and sometimes direct current voltage offsets) are applied between the ring electrode and the two end-cap electrodes establishing a quadrupole-electric field. This field is uncoupled in three directions so that ion motion can be considered independently in each direction; the force acting upon an ion increases with the displacement of the ion from the center of the field but the direction of the force depends on.the instantaneous voltage applied to the ring electrode. A restoring force along one coordinate (such as the distance, r, from the ion-trap's axis of radial symmetry) will exist concurrently with a repelling force along another coordinate (such as the distance, z, along the ion traps axis), and if the field were static the ions would eventually strike an electrode. However, in an RF field the force along each coordinate alternates direction so that a stable trajectory may be possible in which the ions do not strike a surface. In practice, ions of appropriate mass-to-charge ratios may be trapped within the device for periods of milliseconds to hours. A diagram of a typical ion trap is illustrated in Figure" 7. Analysis of stored ions is performed by increasing the RF voltage, which makes the ions successively unstable. The effect of the RF voltage on the ring electrode is to "squeeze" the ions in the xy plane so that they move along the z axis. Half the ions are lost to the top cap (held at ground potential); the remaining ions exit the lower end cap to be detected by the electron multiplier. As the energy applied to the ring electrode is, increased, the ions are collected in order of increasing mass to produce a conventional mass spectrum. With the ion trap, approximately 50 percent of the generated ions are detected. As a result, a significant increase in sensitivity can be achieved when compared to a full scan linear quadrupole system.

7.2.2.4 GC/MS Interface. Any gas chromatograph to mass spectrometer interface that gives acceptable calibration points for each of the analytes of interest and can be used to achieve all acceptable performance criteria may be used. Gas chromatograph to mass spectrometer interfaces constructed of all-glass, glass-lined, or fused silica-lined materials are recommended. Glass.and fused silica should be deactivated.

7.2.2.5 Data System. The computer system that is interfaced to the mass spectrometer must allow the continuous acquisition and storage, on machine readabje media, of all mass spectra obtained throughout the duration of the chromatographic program. The computer must have software that allows searching any GC/MS data file for ions of a specified mass and plotting such ion abundances versus time or scan number. This type of plot is defined as a Selected Ion Current Profile (SICP). Software must also be available that allows integrat­ing the abundance in any SrCP between specified time or scan number limits. Also, software must be available that allows for the comparison of sample spectra with reference library spectra. The National Institute of Standards and Technology (NIST) or Wiley Libraries or equivalent are recommended as reference libraries.

7.2.2.6 Off-line Data Storage Device. Device must be capable of rapid recording and retrieval of data and must be suitable for long-term, off-line data storage.

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-9

VOCs Method TO-15

8.2.2 Pressurized Sampling [see Figure 1 (with metal bellows type pump)]. 8.2.2.1 Pressurized sampling is used when longer-term integrated samples or higher volume samples are

required. The sample is collected in a canister using a pump and flow control arrangement to achieve a typical 101-202 kPa (15-30 psig) final canister pressure. For example, a 6-liter evacuated canister can be tilled at 10 mL/min for 24 hours to achieve a final pressure of 144 kPa (21 psig).

8.2.2.2 In pressurized canister sampling, a metal bellows type pump draws in air from the sampling manifold to till and pressurize the sample canister.

8.2.3 All Samplers. 8.2.3.1 A flow control device is chosen to maintain a constant flow into the canister over the desired

sample period. This flow rate is determined so the canister is filled (to about 88.1 kPa for subatmospheric pressure sampling or to about one atmosphere above ambient pressure for pressurized sampling) over the desired sample period. The flow rate can be calculated by:

F = P x V

T x 60

where:

F = flow rate, mL/min. P = final canister pressure, atmospheres absolute. P is approximately equal to

kPa aauae . is—:=_ + 1

101.2

V = volume of the canister, mL. T = sample period, hours.

For example, if a 6-L canister is to be filled to 202 kPa (2 atmospheres) absolute pressure in 24 hours, the flow rate-can be calculated by:

F = 2 * 6 0 0 0 = 8.3 mL/min 24 x 60

8.2.3.2 For automatic operation, the timer is designed to start and stop the pump at appropriate times for the desired sample period. The timer must also control the solenoid valve, to open the valve when starting the pump and to close the valve when stopping the pump.

8.2.3.3 The use of the Skinner Magnelatch val ve (see Figure 2) avoids any substantial temperature rise that would occur with a conventional, normally closed solenoid valve that would have to be energized during the entire sample period. The temperature rise in the valve could cause outgassing of organic compounds from the Viton® valve seat material. The Skinner Magnelatch valve requires only a brief electrical pulse to open or close at the appropriate start and stop times and therefore experiences no temperature increase. The pulses may be obtained either with an electronic timer that can be programmed for short (5 to 60 seconds) ON periods, or with a conventional mechanical timer and a special pulse circuit. A simple electrical pulse circuit for operating the Skinner Magnelatch solenoid valve with a conventional mechanical timer is illustrated in Figure 2(a). However, with this simple circuit, the valve may operate unreliably during brief power interruptions or if the timer is manually switched on and off too fast. A better circuit incorporating a time-delay relay to provide more reliable valve operation is shown in Figure 2(b).

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-11

'. . Method TO-15

8.3.7 The "practice" canister and certified mass flow meter are disconnected and a clean certified (see Section 8.4.1) canister is attached to the system.

8.3.8 The canister valve and vacuum/pressure gauge valve are opened. 8.3.9 Pressure/vacuum in the canister is recorded on the canister FTDS (see Figure 9) as indicated by the

sampler vacuum/pressure gauge.

8.3.10 The vacuum/pressure gauge valve is closed and the maximum-minimum, thermometer is reset to current temperature. Time of day and elapsed time meter readings are recorded on the canister FTDS.

8.3.11 The electronic timer is set to start and stop the sampling period at the appropriate times. Sampling starts and stops by the programmed electronic timer.

8.3.12 After the desired sampling period, the maximum, minimum, current interior temperature and current ambient temperature are recorded on the FTDS. The current reading from the flow controller is recorded.

.8.3.13 At the end of the sampling period, the vacuum/pressure gauge valve on the sampler is briefly opened and closed and the pressure/vacuum is recorded on the FTDS. Pressure should be close to desired pressure.

[Note: For a subatmospheric sampling system, if the canister is at atmospheric pressure when the field final pressure check is performed, the sampling period may be suspect. This information should be noted on the sampling field data sheet.]

Time of day and elapsed time meter readings are also recorded. 8.3.14 The canister valve is closed. The sampling line is disconnected from the canister and the canister is

removed from the system. For a subatmospheric system, a certified mass flow meter is once again connected to the inlet manifold in front of the in-line filter and a "practice" canister is attached to the Magnelatch valve of the sampling system. The final flow rate is recorded on the canister FTDS (see Figure 9).

[Note: For a pressurized system, the final flow may be measured directly.]

The sampler is turned off.

8.3.15 An identification tag is attached to the canister. Canister serial number, sample number, location, and date, as a minimum, are recorded on the tag. The canister is routinely transported back to the analytical laboratory with other canisters in a canister shipping case.

8.4 Cleaning and Certification Program i

8.4.1 Canister Cleaning and Certification. 8.4.1.1 All canisters must be clean and free of any contaminants before samplecollection. 8.4.1.2 All canisters are leak tested by pressurizing them to approximately 206 kPa (30 psig) with zero

air. . *

[Note: The canister cleaning system in Figure 10 can be used for this task.]

The initial pressure is measured, the canister valve is closed, and the final pressure is checked after 24 hours. If acceptable, the pressure should not vary more than ± 13.8 kPa (± 2 psig) over the 24 hour period.

8.4.1.3 A canister cleaning system may be assembled as illustrated in Figure 10. Cryogen is added to both the vacuum pump and zero air supply traps. The canister(s) are connected to the manifold. The vent shut-off valve and the canister valve(s) are opened to release any remaining pressure in the canister(s). The vacuum pump is started and the vent shut-off valve is then closed and the vacuum shut-off valve is opened. The canister(s

. evacuated to O.05 mm Hg (see Appendix B) for at least I hour. are

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-13

[Note: In the following sections,, "certification" is defined as evaluating the sampling system with humid zero air and humid calibration gases that pass through all active components of the sampling svstem. The svstem is "certified" if no significant additions or deletions'(less than 0.2 ppbv each of target compounds) have occurred when challenged with the test gas stream.]

8.4.3.1 The cleanliness of the sampling system is determined by testing the sampler with humid zero air without an evacuated gas sampling canister, as follows.

8.4.3.2 The calibration system and manifold are assembled, as illustrated in Figure 8. The sampler (without an evacuated gas canister) is connected to the manifold and the zero air.cylinder is~activated to generate a humid gas stream (2 L/min) to the calibration manifold [see Figure 8(b)].

8.4.3.3 The humid zero gas stream passes through the calibration manifold, through the sampling system (without an evacuated canister) to the water management system/VOC preconcentrator of an analyticafsystem.

[Afore; The exit of the sampling system (without the canister) replaces the canister in Figure 11.] •

After the sample volume (e.g., 500 mL) is preconcentrated on the trap, the trap is heated and the VOCs are thermally desorbed and refocussed on a cold trap. This trap is heated and the VOCs are thermally desorbed onto the head of the capillary column. The VOCs are refocussed prior to gas chromatographic separation. Then, the oven temperature (programmed) increases and the VOCs begin to elute and are detected by a GC/MS (see Section 10) system. The analytical system should not detect greater than 0.2 ppbv of any targeted VOCs in order for the sampling system to pass the humid zero air certification test. .Chromatograms (using an FID) of a certified sampler and contaminated sampler are illustrated in Figures 12(a) and 12(b), respectively^ If the sampler passes the humid zero air test, it is then tested with humid calibration gas standards containing selected VOCs at concentration levels expected in field sampling (e.g., 0.5 to 2 ppbv) as outlined in Section 8.4.4.

8.4.4 Sampler System Certification with Humid Calibration Gas Standards from a Dynamic Calibration System

8.4.4.1 Assemble the dynamic calibration system and manifold as illustrated in Figure 8. ' 8.4.4.2 Verify that the calibration system is clean (less than 0.2 ppbv of any target compounds) by

sampling a humidified gas stream, without gas calibration standards, with a previously certified clean canister (see Section 8.1).

8.4.4.3 The assembled dynamic calibration system is certified clean if less than 0.2 ppbv of any targeted compounds is found.

8.4.4.4 For generating the humidified calibration standards, the calibration gas cylinder(s) containing nominal concentrations of 10 ppmv in nitrogen of selected VOCs is attached to the calibration system as illustrated in Figure 8. The gas cylinders are opened and the gas mixtures are passed through 0 to 10 mL/min certified mass flow controllers to generate ppb levels of calibration standards.

8.4.4.5 After the appropriate equilibrium period, attach the sampling system (containing a certified evacuated canister) to the manifold, as illustrated in Figure 8(b).

8.4.4.6 Sample the dynamic calibration gas stream with the sampling system. 8.4.4.7 Concurrent with the sampling system operation, .realtime monitoring of the calibration gas stream

is accomplished by the on-line GC/MS analytical system [Figure 8(a)] to provide reference concentrations of generated VOCs.

8.4.4.8 At the end of the sampling period (normally the same time period used for experiments), the sampling system canister is analyzed and compared to the reference GC/MS analytical system to determine if the concentration of the targeted VOCs was increased or decreased by the sampl ing system.

8.4.4.9 A recovery of between 90% and 1 10% is expected for all targeted VOCs. 8.4.5 Sampler System Certification without Compressed Gas Cylinder Standards.

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-15

VOCs Method TO-15

9.2 Preparation of Standards

9.2.1 Introduction. 9.2.1.1 When available, standard mixtures of target gases in high pressure cylinders must be certified

traceable to a NIST Standard Reference Material (SRM) or to a NIST/EPA approved Certified Reference Material (CRM). Manufacturer's certificates of analysis must be retained to track the expiration date.

9.2.1.2 The neat standards that are used for making trace gas standards must be of high purity; generally a purity of 98 percent or better is commercially available.

9.2.1.3 Cylinder(s) containing approximately 10 ppmv of each of the target compounds are typically used as primary stock standards. The components may be purchased in one cylinder or in separate cylinders depending on compatibility of the compounds and the pressure of the mixture in the cylinder. Refer to manufacturer's specifications for guidance on purchasing and mixing VOCs in gas cylinders.

9.2.2 Preparing Working Standards. 9.2.2.1 Instrument Performance Check Standard. Prepare a standard solution of BFB in humidified

zero air at a concentration which will allow collection of 50 ng of BFB or less under the optimized concentration parameters.

9.2.2.2 Calibration Standards. Prepare five working calibration standards in humidified zero air at a concentration which will allow collection at the 2, 5, 10, 20, and 50 ppbv level for each component.under the optimized concentration parameters.

9.2.2.3 Internal Standard Spiking Mixture. Prepare an internal spiking mixture containing bromo-chloromethane, chlorobenzene-d5, and 1,4-difluorobenzene at 10 ppmv each in humidified zero air to be added to the sample or calibration standard. 500 uL of this mixture, spiked into 500 mL of sample will result in a concentration of 10 ppbv. The internal standard is introduced into the trap during the collection time for all calibration, blank, and sample analyses using the apparatus shown in Figure 13 or by equivalent means. The volume of internal standard spiking mixture added for each analysis must be the same from run to run.

9.2.3 Standard Preparation by Dynamic Dilution Technique. 9.2.3.1 Standards may be prepared by dynamic dilution of the gaseous contents of a cylinder(s) containing

the gas calibration stock standards with humidified zero air using mass flow controllers and a calibration manifold. The working standard may be delivered from the manifold to a clean, evacuated canister using a pump and mass flow controller.

9.2.3.2 Alternatively, the analytical system may be calibrated by sampling directly from the manifold if the flow rates are optimized to provide the desired amount of calibration.standards. However, the use of the canister as a reservoir prior to introduction into the concentration system resembles the procedure normally used to collect samples and is preferred. Flow rates of the dilution air and cylinder standards (all expressed in the same units) are measured using a bubble meter or calibrated electronic flow measuring device, and the concentrations of target compounds in the manifold are then calculated using the dilution ratio and the original concentration of each compound.

9.2.3.3 Consider the example of 1 mL/min flow of 10 ppmv standard diluted with 1,000 mL/min of humid air provides a nominal 10 ppbv mixture, as calculated below:

Manifold Cone. = (Original >Conc.) (Std. Gas Flowrate) (Air Flowrate) + (Std. Gas Flowrate)

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-17

VOCs Method TO-15

9.2.5.2 An aluminum cylinder is flushed with high-purity nitrogen gas and then evacuated to better than 25 in. Hg.

9.2.5.3 Predetermined amounts of each neat standard compound are measured using a microliter or gastight syringe and injected into the cylinder. The cylinder is equipped with a heated injection port and nitrogen flow to facilitate sample transfer.

9.2.5.4 The cylinder is pressurized to 1000 psig with zero nitrogen.

[Note: User should read all SOPs associated with generating standards in high pressure cylinders. Follow all safety requirements to minimize danger from high pressure cylinders.]

9.2.5.5 The contents of the cylinder are allowed to equilibrate (-24 hrs) prior to withdrawal of aliquots into the GC system.

9.2.5.6 If the neat standard is a gas, the cylinder concentration is determined using the following equation:

Volume , , Concentration, ppbv = s o n d a r d x 109

• V 0 l l I m e d i l u u o n gns

[Note: Both values must be expressed in the same units.]

9.2.5.7 If the neat standard is a liquid, the gaseous concentration can be determined using the following equations:

y - n R T

and:

n = (mL)(d) MW

where: V = Gaseous volume of injected compound at EPA standard temperature (25°C) and pressure (760 mm Hg), L.

n= Moles. R = Gas constant, 0.08206 L-atm/mole °K. T = 298 ° K (standard temperature). P = 1 standard pressure, 760 mm Hg (1 atm).

mL= Volume of liquid injected, mL. d = Density of the neat standard, g/mL.

MW = Molecular weight of the neat standard expressed, g/g-mole.

The gaseous volume of the injected compound is divided by the cylinder volume at STP and then multiplied by 10'J to obtain the component concentration in ppb units.

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-19

VOCs Method TO-15

Set point Sample volume Carrier gas purge flow

-150°C -up to 100 mL - none •

Set point Sample volume Carrier gas purge flow

27°C - up to 1,000 mL - selectable

[Note: The analyst should optimize the flow rate, duration of sampling, and absolute sample volume to be used. Other preconcentration. systems may be used provided performance standards (see Section 11) are realized.]

10.1.2 Desorption Conditions

Cryogenic Trap

Desorb Temperature Desorb Flow Rate Desorb Time

120°C - 3 mL/min He <60 sec

Adsorbent Trap

Desorb Temperature Desorb Flow Rate Desorb Time

Variable ~3 mL/min He <60 sec

The adsorbent trap conditions depend on the specific solid adsorbents chosen (see manufacturers' specifications).

10.1.3 Trap Reconditioning Conditions.

Cryogenic Trap

Initial bakeout Variable (24 hrs) After each run

10.2 GC/MS System

120°C(24hrs)

120°C(5 min)

Adsorbent Trap

Initial bakeout

After each run Variable (5 min)

• 10.2.1 Optimize GC conditions for compound separation and sensitivity. Baseline separation of benzene and carbon tetrachloride on a 100% methyl pofysiloxane stationary phase is an indication of acceptable chromatographic performance.

10.2.2 The following are the recommended gas chromatographic analytical conditions when using a 50-meter by 0.3-mm I.D., 1 um film thickness fused silica column with refocusing on the column.

Item

Carrier Gas: Flow Rate: Temperature Program:

Condition

Helium Generally 1-3 mL/min as recommended by manufacturer Initial Temperature: -50 °C Initial Hold Time: 2 min Ramp Rate: 8° C/min Final Temperature: • 200°C Final Hold Time: Until all target compounds elute.

10.2.3 The following are the recommended mass spectrometer conditions:

Item Condition

I Is" I m I Ha ! IE

I m

-w - IE. i E . If)

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-21

VOCs Method TO-15

10.4.6 Documentation. Results of the BFB tuning are to be recorded and maintained as part of the instrumentation log.

10.5 Initial Calibration

.' 10.5.1 Summary. Prior to the analysis of samples and blanks but after the instrument performance check standard criteria have been met, each GC/MS system must be calibrated at five concentrations that span the monitonng range ot interest in an initial calibration sequence to determine instrument sensitivitv and the linearitv of GC/MS response for the target compounds. For example, the range of interest mav be 2 to 20 ppbv in- which case the five concentrations would be 1, 2, 5, 10 and 25 ppbv.

One of the calibration points from the initial calibration curve must be at the same concentration as the dailv calibration standard (e.g., 10 ppbv).

10.5.2 Frequency. Each GC/MS system must be recalibrated following corrective action (e a ion source cleaning or repair, column replacement, etc.) which may change or affect the initial calibration crite'ria or if the daily calibration acceptance criteria have not been met.

If time remains in the 24-hour time period after meeting the acceptance criteria for the initial calibration samples may be analyzed.

If time does not remain in the 24-hour period after meeting the acceptance criteria for the initial calibration a new analytical sequence shall commence with the analysis of the instrument performance check standard followed by analysis of a daily calibration standard.

10.5.3 Procedure. Verify that the GC/MS system meets the instrument performance criteria in Section 10.4.

The GC must be operated using temperature and flow rate parameters equivalent to those in Section 10 ? 2 Calibrate the preconcentration-GC/MS system by drawing the standard into the system.. Use one of the standards preparation techniques described under Section 9.2 or equivalent.

A minimum of five concentration levels are needed to determine the instrument sensitivity and linearity One of the calibration levels should be near the detection level for the compounds of interest. The calibration range should be chosen so that linear results are obtained as defined in Sections 10.5.1 and 10.5.5.

Quantitation ions for the target compounds are shown in Table 2. The primary ion should be used unless interferences are present, in which case a secondary ion is used.

10.5.4 Calculations.

[Note: In the following calculations, an internal standard approach is used to calculate response factors The area response used is that of the primary quantitation ion unless otherwise stated.]

10.5.4.1 Relative'Response Factor (RRF). Calculate the relative response factors for each target compound relative to the appropriate internal standard (i.e., standard with the nearest retention time) usin* the following equation: a

• A C RRF = x i s

January 1999 Compendium of Methods forToxic Organic Air Pollutants Page 15-23

VOCs Method TO-15

RRT = £ H I i=i n

where: RRT - Mean relative retention time for the target compound for each initial calibration standard.

RRT = Relative retention time for the target compound at each calibration"leVel. 10.5.4.6 Tabulate Primary Ion Area Response (Y) for Internal Standard. Tabulate the area response

(Y) of the primary ions (see Table 2) and the corresponding concentration for each compound and internal standard.

10.5.4.7 Mean Area Response (Y) for Internal Standard. Calculate the mean area response (Y) for each internal standard compound over the initial calibration range using the following equation:

n v

Y = £ l i ; i=i n

where: Y = Mean area response. ,.

Y = Area response for the primary quantitation ion for the internal standard for each initial calibration standard.

10.5.4.8 Mean Retention Times (RT). Calculate the mean of the retention times (RT) for each internal standard over the initial calibration range using the following equation:

RT = J2 1

i=i n

where: RT = Mean retention time, seconds

RT = Retention time for the internal standard for each initial calibration standard, seconds: 10.5.5 Technical Acceptance Criteria for the Initial Calibration.

10.5.5.1 The calculated %RSD for the RRF for each compound in the calibration table must be less than 30% with at most two exceptions up to a limit of 40%,

[Note: This exception may not be acceptable for all projects. Many projects may have a specific target list of compounds which would require the lower limit for all compounds.]

10.5.5.2 The RRT for each target compound at each calibration level must be withiin 0.06 RRT units of the mean RRT for the compound.

10.5.5.3 The area response Y of at each calibration level must be within 40% of the mean area response Y over the initial calibration range for each internal standard.

10.5.5.4 The retention time shift for each of the internal standards at each calibration level must be within 20 s of the mean retention time over the initial calibration range for each internal standard.

10.5.6 Corrective Action. 10.5.6.1 Criteria. If the initial calibration technical acceptance criteria are not met, inspect the system

for problems. It may be necessary to clean the ion source, change the column, or take other corrective actions to meet the initial calibration technical acceptance criteria..

10.5.6.2. Schedule. Initial calibration acceptance criteria must be met before any field samples, performance evaluation (PE) samples, or blanks are analyzed.

*! It! IE

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-25

VOCs • '• : Method TO-15

usmg a,, reagents, standards, equipment, apparatus, glassware, and solvents that would be used for a sample

A laboratory method blank (LMB) is an unused, certified canister that has not left the laboratory Th, hi L canister is pressurized with humidified ultra-mire rem air ,„H • A , moratory. The blank

10.7.2 Frequency. The laboratory method blank must be analyzed after the calihnrinn « A A, s A before any samples are analyzed. • calibration standard(s) and

under Section 10.8.

apply equations in Section 10.5.4

The blank sample should be analyzed using the same procedure outlined U 1 I U C I o c l

10.7.4 Calculations. The blanks are analyzed similar to a field sample and the

10.7.5 Technical Acceptance Criteria. A blank canister should be analyzed daily.

The retention time for each of the internal standards must be within ±0 3 most recent valid calibration. 3 minutes between the blank and the

10.8 Sample Analysis

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-27

VOCs Method TO-15

A, = Area of the characteristic ion tor the compound to be measured, counts.

A i s = Area of the characteristic ion for the specific internal standard, counts.

Cis = Concentration of the internal standard spiking mixture, ppbv

RRF - Mean relative response factor from the initial calibration.

DF = Dilution factor calculated as described in section 2. If no dilution is performed. DF = 1.

[More: The equation above is valid under the condition that the volume (-500 pi) of internal standard spiking mixture added in all field and QC analyses is the same from run to run, and that the volume (~ 500 mL) of held and QC sample introduced into the trap is the same for each analysis.]

10.8.5 Technical Acceptance Criteria.

(Note: If the most recent valid calibration is an initial calibration, internal standard area responses and RTs in the sample are evaluated against the corresponding internal standard area responses and RTs in the mid level standard (10 ppbv) of the initial calibration.] •

10.8.5.1 The field sample must be analyzed on a GC/MS svstem meeting the BFB tuning initial calibration, and continuing calibration technical acceptance criteria at the frequency described in Sections 10 4 10.5 and 10.6. ' '

10.8.5.2 The field samples-must be analyzed along with,a laboratory method blank that met the blank technical acceptance criteria.

10.8.5.3 All of the target analyte peaks should be within the initial calibration range. 10.8.5.4 The retention time for each internal standard must be within ±0.33 minutes of the retention time

ot the internal standard in the most recent valid calibration. 10.8.6 Corrective Action. If the on-column concentration of any compound in any sample exceeds the

initial calibration range, an aliquot of the original sample must be diluted and reanalvzed. Guidance in performing dilutions and exceptions to this requirement are given below.

• Use the results of the original analysis to determine the approximate dilution factor required to *et the largest analyte peak within the initial calibration range. &

• The dilution factor chosen should keep the response of the largest analyte peak for a target compound in the upper half of the initial calibration range of the instrument.

[Note: Analysis involving dilution should be reported with a dilution factor and nature of the dilution gas.]

10.8.6.1 Internal standard responses and retention times must be evaluated during or immediately after data acquisition. If the retention time for any internal standard changes by more than 20 sec from the latest daily (24-hour) calibration standard (or mean retention time over the initial calibration range), the GC/MS system must be inspected tor malfunctions, and corrections made as required.

10.8.6.2 If the area response for any internal standard changes bv more than ±40 percent between the sample and the most recent valid calibration, the GC/MS system must be inspected tor malfunction and

Ho •ft! :ie IE

I It!

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-29

VOCs Method TO-15

11.3.2 There are several factors which may affect the precision of the measurement. The nature of the compound of interest itself such as molecular weight, water solubility, polarizability, etc., each have some effect on die precision, for a given sampling and analytical system. For example, styrene, which is classified as a polar VOC, generally shows slightly poorer precision than the bulk of nonpolar VOCs. A primary influence on precision is the concentration level of the compound of interest in the sample, i.e., the precision degrades as the concentration approaches the detection limit. A conservative measure was obtained from replicate analysis of "real world" canister samples from the TAMS and UATMP networks. These data are summarized in Table 5 and suggest that a replicate precision value of 25 percent can be achieved for each of the target compounds.

11.4 Audit Accuracy

11.4.1 A measure of analytical accuracy is the degree of agreement with audit standards. Audit accuracy is defined as the difference between the nominal concentration of the audit compound and the measured value divided by the.audit value and expressed as a percentage, as illustrated in the following equation:

A J V A 0 / Spiked Value - Observed Value Audit Accuracy, % = —!- x 100 Spiked Value

11.4.2 Audit accuracy results for TAMS and UATMP analyses are summarized in Table 6 and were used to form the basis for a selection of 30 percent as the performance criterion for audit accuracy.

i A

i -.

12. References

1. Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-14A, Second Edition, U. S. Environmental Protection Agency, Research Triangle Park, NC, EPA 600/625/R-96/010b, January 1997.

2. Winberry, W. T, Jr., et al., Statement-of-Work (SOW) for the Analysis of Air Toxics From Superfund Sites, U. S. Environmental Protection Agency, Office of Solid Waste, Contract Laboratory Program, Washington, D.C., Draft Report, June 1990.

3. Coutant, R.W., Theoretical Evaluation of Stability of Volatile Organic Chemicals and Polar Volatile Organic Chemicals in Canisters, U. S. Environmental Protection Agency, EPA Contract No. 68-DO-0007, Work Assignment No. 45, Subtask 2, Battelle, Columbus, OH, June 1993.

4. Kelly, T.J., Mukund, R., Gordon, S.M., and Hays, M.J., Ambient Measurement Methods and Properties of the 189 Tide I I I Hazardous Air Pollutants, U. S. Environmental Protection Agency, EPA Contract No. 68-DO-0007, Work Assignment 44, Battelle, Columbus, OH, March 1994.

5. Kelly T. J. and Holdren, M.W., "Applicability of Canisters for Sample Storage in the Determination of Hazardous Air Pollutants," Atmos. Environ., Vol. 29, 2595-2608, May 1995.

6. Kelly, T.J., Callahan, P.J., Pleil, J.K., and Evans, G.E., "Method Development and Field Measurements for Polar Volatile Organic Compounds in Ambient Air," Environ. Sci. Techno!., Vol. 27, 1146-1153, 1993. '

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 15-31

7. McCIenny, W.A.. Oliver, K..D. and Daughtrey, E.H.., Jr. "Dry Purging of Solid Adsorbent Traps to Remove Water Vapor Before Thermal Desorption of,Trace Organic Gases," J. Air and Waste Manag. Assoc., Vol 45, 792-800, June 1995.

8. Whitaker, D.A., Fortmann, R.C. and Lindstrom, A.B. "Development and Testing of a Whole Air Sampler for Measurement of Personal Exposures to Volatile Organic Compounds," Journal of Exposure Analysis and Environmental Epidemiology, Vol. 5, No. 1, 89-100, January 19.95. '

9. Pleil. J.D. and Lindstrom, A.B., "Collection of a Single Alveolar Exhaled Breath for Volatile Organic Compound Analysis," American Journal of Industrial Medicine,,.Vol. 28, 109-121, 1995.

. 10. Pleil, J.D. and McCIenny, W.A., "Spatially Resolved Monitoring for.Volatile Organic Compounds Using Remote Sector Sampling," Atmos. Environ., Vol. 27A, No. 5, 739-747, August 1993.

11. Holdren, M.W., et al., Unpublished Final Report, EPA Contract 68-DO-0007, Battelle, Columbus, OH. Available from J.D. Pleil, MD-44, U. S. Environmental Protection Agency, Research Trianale Park. NC, 27711, 919-541-4680.

12. Morris, CM., Burkley, R.E. and Bumgarner, J.E., "Preparation of Multicomponent Volatile Organic Standards Using Dilution Bottles," Anal. Letts., Vol. 16 (A20), 1585-1593, 1983.

13. Pollack, A.J., Holdren, M.W., "Multi-Adsorbent Preconcentration and Gas Chromatographic Analysis of Air Toxics With an Automated Collection/Analytical System," in the Proceedings of the 1990 EPA/A&WMA International Symposium of Measurement of Toxic and Related Air Pollutants, U. S. Environmental Protection Agency, Research Triangle Park, NC, EPA/600/9-90-026, May 1990.

14. Stephenson, J.H.M., Allen, F., Slagle, T., "Analysis of Volatile Organics in Air via Water Methods" in Proceedings of the 1990 EPA/A&WMA International Symposium on Measurement of Toxic and Related Air Pollutants, U. S. Environmental Protection Agency, Research Triangle Park, NC, EPA 600/9-90-026, May 1990.

15. Oliver, K. D., Adams, J. R., Davehtrey, E. H., Jr.; McCIenny, W. A., Young, M. J., and Parade, M. A., "Techniques for Monitoring Toxices VOCs in Air: Sorbent Preconcentration Closed-Cycle Cooler Cryofocusing, and GC/MS Analysis," Environ. Sci. Techno!., Vol. 30, 1938-1945, 1996.

Page 15-32 Compendium of Methods for Toxic Organic Air Pollutants January 1999

APPENDIX B.

COMMENT ON CANISTER CLEANING PROCEDURES

The canister cleaning procedures given in Section 8.4 require that canister pressure be reduced to <0.05mm Hg before the cleaning process is complete. Depending on the vacuum system design (diameter of connecting tubing, valve restrictions, etc.) and the placement of the vacuum gauge, the achievement of this value'may take several hours. In any case, the pressure gauge should be placed near the canisters to determine pressure. The objective of requiring a low pressure evacuation during canister cleaning is to reduce contaminants. If canisters can be routinely certified (<0.2 ppbv for target compounds) while using a higher vacuum, then this criteria can be relaxed. However, the ultimate vacuum achieved during cleaning should always be <0.2mm Hg.

Canister cleaning as described in Section 8.4 and illustrated in Figure 10 requires components with special features. The vacuum gauge shown in Figure 10 must be capable of measuring 0.05mm Hg with less than a 20% error. The vacuum pump used for evacuating the canister must be noncontaminating while being capable of achieving the 0.05 mm Hg vacuum as monitored near the canisters. Thermoelectric vacuum gauges and turbomolecular drag pumps are typically being used for these two components.

An alternate to achieving the canister certification requirement of <0.2 ppbv for all target compounds is the criteria used in Compendium Method TO-12 that the total carbon count be <10ppbC. This check is less expensive and typically more exacting than the current certification requirement and can be used if proven to be equivalent to the original requirement. This equivalency must be established by comparing the total nonmethane organic carbon (TNMOC) expressed in ppbC to the requirement that individual target compounds be <0.2 ppbv for a series of analytical runs.

Page 15-34 Compendium of Methods for Toxic Organic Air Pollutants January 1999

Method TO-15 VOCs

APPENDIX D.

LISTING OF COMMERCIAL SUPPLIERS OF PERMEATION TUBES AND SYSTEMS

Kin-Tek 504 Laurel St. Lamarque. Texas 77568 (409) 938-3627 (800) 326-3627

Vici Metronics, Inc. 2991 Corvin Drive Santa Clara, C A 95051 (408)737-0550

Analytical Instrument Development. Inc. Rt. 41 and Newark Rd. Avondale, PA 19311 (215) 268-3181

Ecology Board, Inc. 9257 Independence Ave. Chatsworth, CA 91311 (213)882-6795

Tracor, Inc. . • . . -Jo. 6500 Tracor Land m Austin, TX • « (512) 926-2800 ' |S

Metronics Associates,.Inc. J ^ 3201 Porter Drive / Standford Industrial Park !,: Palo Alto, CA 94304 . p (415)493-5632

Page 15-36 Compendium of Methods for Toxic Organic Air Pollutants January 1999

TABLE 1. (continued)

Compound CAS No. BP (°C) y.p.

(mmllg)1 M W TO-I4A CLP-SOW Chloroprene (2-chloro-l ,3-butadiene); C4H5CI 126-99-8 59.4 226 88.5 Chloromelhyl methyl ether; C2H5CIO 107-30-2 59.0 224 805 Acrolein (2-propenal); C3H40 107-02-8 52.5 220 56 X 1,2-Epoxybutane (1,2-butylene oxide); C4H80 106-88-7 63.0 163 72 Chloroform; CHCI3 ! 67-66-3 61.2 160 119 X X Ethyleneimine (aziridine); C2H5N ' 151-56-4 56 160.0 43 .1,1-Dimethylhydrazine; C2H8N2 57-14-7 63 157.0 60.0 Hexane;C6HI4 110-54-3 69.0 120 86.2 X 1,2-Propyleneimine (2-methylaziridine); C3H7N 75-55-8 66.0 112 57.1 Acrylonitrile (2-propenenitrile); C3H3N 107-13-1. 77.3 100 53 X Methyl chloroform (1,1,1 -trichloroethane); C2H3CI3 71-55-6 74.1 100 133.4 X - X Methanol; CH40 67-56-1 65.0 92.0 32 X Carbon tetrachloride; CCI4 56-23-5 76.7 90.0 153.8 X X Vinyl acetate; C4H602 108-05-4 72.2 83.0 86 X Methyl ethyl ketone (2-butanone); C4H80 . - - 78-93-3 79.6 77.5 72 X Benzene; C6H6 71-43-2 80.1 76.0 78 X X Acetonitrile (cyanomethane); C2H3N 75-05-8 82 74.0 41.0 X Ethylene dichloride (1,2-dichloroethanc); C2I I4CI2 107-06-2 • 83.5 61.5 99 X X Triethylamine; C6H15N 121-44-8 89.5 54.0 101.2 Methylhydrazine; CH6N2 60-34-4 87. 8 49.6 46.1

Propylene dichloride (1,2-dichloropropane); C3H6CI2 • 78-87-5 97.0 42.0 113 X X 2,2,4-Ti imethyl pentane CSHI8 540-84-1 99.2 40.6 114 1,4-Dioxane (1,4-Diethylene oxide); C4H802 "' " 123-91-1 101 37.0 ' 88 Bis(chloromethyl) ether; C2H4CI20 542-88-1 104 30.0 115 Ethyl acrylate; C5H802 140-88-5 100 29.3 100

Methyl methacrylate; C5H802 80-62-6 101 28.0 100.1

Compound

Cumene (isopropylbenzene); C9H 12

Acrylic acid; C3H402

N.N-Dimethylformamide: C3H7NO

1,3-Piopane sultone; C3H6Q3S

Acetophenone; CSII80

Dimethyl sulfate; C2H604S

TABLE 1. (continued)

CAS No.

98-82-8

79-10-7

68-12-2

1120-71-4

98-86-2

77-78-

BP(°C)

153

153

180/30n

202

. v.p. (mmllg)'

3.2

3.2

2.7

2.0

1.0

1.0

M W

120

72

TO-I4A

73

122.1

120

126.1

CLP-SOW

Benzyl chloride (a-chlorotoluene); C7H7CI

,1,2-Dibrotno-3-chloropropane; C3H5Br2CI

Bis(2-Chloroethyl)ether; C4H8C120

Chloroacetic acid; C2H3CIQ2

Ariiline (aminobenzene); C6H7N

,4rDichlorobenzcne (p-); C6H4CI2

Ethyl carbamate (methane); C3H7NQ2

Aciylamide; C3H5NO

N,N-Dimethylaniline; C8HI IN

Hexachloroethane; C2CI6

lexachlotohutadicne; C4CI6

sophorone; C9HI4Q

N-Nitrosomorpholine; C4H8N2Q2

Stymie oxide; CSH80

Diethyl sulfate; C4H10O4S

Cresylic acid (cresol isomer mixture);C7H8Q

o-Cresol; C7H80

Catechol (o-hydroxyphenol); C6H6Q2

Phenol; C6H6Q

100-44-7 179 1.0 126.6

96-12-8 196 0.80 236.4

111-44-4 178 0.71 143

79-11-8 189 0.69 94.5

62-53-3 0.67

106-46-7 173 0.60

51-79-6 183

79-06-1 125/25 mm

0.54

0.53

93

147

89

71

121-69-7 192 0.50 121 67-72- Sublimes at 186 0.40 236.7

87-68-3 215 0.40 260.8

78-59-1 215 0.38 138.2

59-89-2 225 0.32 I 16.

96-09-3 194 0.30 120.2 64-67-5 208 0.29 154

1319-77-3 202 0.26 108

95-48-7 191 0.24 108

120-80-9 240

108-95-2 182

0.22

0.20 94

Method TO-15 VOCs

TABLE 2. CHARACTERISTIC MASSES (M/Z) USED FOR QUANTIFYING THE TITLE III CLEAN AIR ACT AMENDMENT COMPOUNDS

Compound CAS No. Primary Ion Secondary Ion Methyl chloride (chloromethane): CH3C1 74-87-3 50 52 Carbonvl sulfide: COS 463-S8-I 60 62 Vinvl chloride (ehloroethene): C2H3C! 7S-01-4 62 64 Diazomethane: CH2N2 334-88-3 42 41. Formaldehyde: CH20 50-00-0 29 30 1.3-Butadiene: C4H6 106-99-0 39 54 Methyl bromide (bromomethane): CH3Br 74-83-9 94 96 Phoscene: CC120 75-44-5 63 65 Vinyl bromide (bromoethene): C2H3Br 593-60-2 106 108 Ethvlene oxide: C2H40 75-21-8 29 44 Ethyl chloride (chloroethane): C2H5C1 75-00-3 64 66 Acetaldehyde (ethanal): C2H40 75-07-0 44 29. 43 Vinylidene chloride 1 1.1-dichloroethvlene): C2H2C12 75-35-4 61 ' 96 Propylene oxide: C3H60 ' 75-56-9 58 57 Methyl iodide (iodomethane): CH3I 74-88-4 142 127 Methylene chloride: CH2C12 75-09-2 49 84. 86 Methvl isocvanate: C2H3NO 624-83-9 .57 56 Allyl chloride (3-chloropropene): C3H5C1 107-05-1 76 41.78 Carbon disulfide: CS2 75-15-0 76 44. 78 Methyl tert-butyl ethen C5H120 1634-04-4 73 ' 41. 53 Propionaldehyde: C2H5CHO 123-38-6 58 29. 57 Ethylidene dichloride (1.1-dichloroethane): C2H4C12 . 75-34-3 63 65. 27

Chloroprene (2-chloro-1.3-butadiene): C4H5C1 126-99-8 . 88 53. 90 Chloromethvl methvl ether: C2H5C10 107-30-2 45 29. 49 Acrolein (2-propenal); C3H40 107-02-8 56 55

1.2-Epoxybutane (1.2-butvlene oxide); C4H80 106-88-7 42 41. 72 Chloroform: CHC13 67-66-3 83 85. 47 Ethyleneimine (aziridine): C2H5N 151-56-4 42 43 1.1-Dimethylhydrazine: C2H8N2 57-14-7 60 45. 59 Hexane:C6H14 110-54-3 57 41. 43 1.2-Propyleneimine (2-methvlazindine): C3H7N 75-55-8 56 ' 57:42 Acrylonitrile (2-propenenitrile): C3H3N 107-13-1 53 52 Methyl chloroform (1.1.1 trichloroethane): C2H3CI3 71-55-6 •97 • 99.61 Methanol: C1-140 67-56-1 31 29 Carbon tetrachloride: CC14 56-23-5 117 1 19 Vinvl acetate: C4H602 108-05-4 43 86 Methvl ethvl ketone (2-butanone): C4I-I80 78-93-3 43 72

Page 15-42 Compendium of Methods for Toxic Organic Air Pollutants January 1999

Method TO-15 VOCs

Compound CASNo. Primarv Ion Secondary Ion Aeetophenone: C8HSO 98-86-2 105 . 77.120 Dimethyl sulfate: C2H604S . 77-78-1 95 66.96 Benzyl chloride (a-chlorotoluene): C7H7CI 100-44-7 91 126 l.2-Dibromo-3-chloroDroDane: C3H5Br2CI 96-12-8 57 155. 157 Bis(2-Chloroethvl)ether: C4HSCI20 111-44-4 93 63. 95 Chloroacetic acid: C2H3CI02 79-11-8 50 45. 60 Aniline (aminobenzene): C6H7N 62-53-3 93 66 1.4-Dichlorobenzene (p-): C6H4CI2 106-46-7 146 148. i 1 1 Ethyl carbamate (urethane): C3H7N02 51-79-6 31 44. 62 Acrylamide: C3H5NO 79-06-1 .44 55. 71 N.N-Dimethvlaniline: C8H1 IN 121-69-7 120 77. 121 He.xachloroethane: C2CI6 67-72-1 201 199. 203 Hexachlorobutadiene: C4CI6 87-68-3 225 227. 223 Isophorone: C9HI40 7S-59-1 82 138 N-Nitrosomorpholine: C4H8N202 :

59-89-2 56 86. 1 16 Styrene oxide: C8H80 96-09-3 • 91 120 Diethyl sulfate: C4H10O4S 64-67-5 45 59. 139 Cresylic acid (cresol isomer mixture); C7H80 . 1319-77-3 o-Cresol: C7H80 95-48-7 108 107 Catechol (o-hydroxvphenol): C6H602 120-80-9 110 64 Phenol: C6H60 108-95-2 94 66 1.2.4-Trichlorobenzene: C6H3CI3 120-82-1 180 182. 184 Nitrobenzene: C6H5N02 ]; 98-95-3 77 ' 51. 123

Page 15-44 Compendium of Methods for Toxic Organic Air Pollutants Ja

Method TO-15 VOCs

TO-l4AList Lab#k SCAN Lab #2. SIM Benzene 0.34 0.29 Benzyl Chloride

Carbon tetrachloride 0.42 0.15 Chlorobenzene 0.34 ' 0.02 Chloroform ' 0.25 0.07 1.3-Dichlorobenzene • 0.36 0.07 1.2-Dibromoethane 0.05 1.4-Dichlorobenzene 0.70 0.12 1.2-Dichlorobenzene 0.44 1.1-Dichloroethane 0.27 0.05 1.2-Dichloroethane 0.24 l.l-Dichloroethene 0.22 cis-1.2-Dichloroethene 0.06 Methvlene chloride 1.38 0.84 1.2-DichloroDrooane 0.21 cis-1.3-DichloroDTODene : 0.36 trans-1.3-Dichloroprooene , 0.22 Ethvlbenzene 0.27 0.05 Chloroethane 0.19 Trichlorofluoromethane

1.1.2-Trichloro-1.2.2-trifluoroethane

1.2-Dichloro-1.1.2.2-tetrafluoroethane

Dichlorodifluoromethane

Hexachlorobutadiene

Bromomethane ' 0.53 Chloromethane 0.40 Stvrene 1.64 0.06 1.1 -2.2-Tetrachloroethane 0.28 0.09 Tetrachloroethene 0.75 0.10 Toluene 0.99 0.20 1.2.4-Trichlorobenzene

l.I.I-Trichloroethane 0.62 0.21 1.1.2-Trichloroethane . 0.50 Trichloroethene 0.45 0.07 1.2.4-Trimethvlbenzene

I.3.5-Trimethvlbenzene

Vinvl Chloride 0.33 0.48 m.p-Xvlene 0.76 0.08 o-Xylene 0.57 0.28

« •Ha IE IE

«n, IO me (JIUUUCI oi tne stanaarci deviation ot seven replicate analyses and the student's "t" test value for 99% confidence. For Lab #2, the MDLs represent an average over four studies MDLs are for MS/SCAN for Lab ft I and for MS/SIM for Lab W>

Page 15-46 Compendium of Methods for Toxic Organic Air Pollutants . January 1999

Method TO-15 VOCs

To AC

Inlet

Insulated Enclosure

Electronic Timer

Electronic Timer

Electronic Timer

Vocuum/Pressure Gouge

Valve

To AC

Figure 1. Sampler configuration for subatmospheric pressure or pressurized canister sampling. '

Page 15-48 Compendium of Methods for Toxic Organic Air Pollutants January 1999

Method TO-15 VOCs

Heated Enclosure

Inlet

A

~ 1 . 6 Meters (~5 f t )

Ground Level

7T

Inlet Manifold

Pump

Thermostat

Prsssura Gouge

T Vent

Electronic Timer

Heater

Tan

I ; _ _ _ _

I Vent

T

I Auxi l iary Vacuum

Pump

Vacuum/Pressure Gauge

Magnelatch Volve

Vent

r Valve)

Canister

1

To AC

Figure 3. Alternative sampler configuration for pressurized canister sampling.

Page 15-50 Compendium of Methods for Toxic Organic Air Pollutants January 1999

o

Z D O O

Si tr < LU W z o to uu rr z LU CD O CC Q > X

150

140

130

120

110 -

1 0 0 , r

100 200 300 400 500 600 700 800 900 1000 1100 PURGE VOLUME, ml

Figure 5. Residua! water vapor on VOC concentrator vs. dry He purge volume.

Colibrotion Gos Cylinder

Zero Air Cylinder

(o) Reol Time GC-FID-ECD-PID

or GC-MS

Moss Flow Control ler

( 0 - 5 0 m L / m i n )

Mass Flow Control ler

( 0 - 5 0 L / m i n )

(b) Evacuated or Pressurized Conister Sampling System

500 mL R o u n d - B o t t o m

Flask

Humidif ier

Vocuum/Pressure Gouge

Shut Off . Volve

(c) Canister Transfer Slandord

Figure 8. Schematic diagram of calibration system and manifold for (a) analytical system calibration, (b) testing canister sampling system and (c) preparing canister transfer standards

Pressure Regulator

Exhaust

Exhaust

Vocuum Pump Vent Shut. Off Valve Valve Check Valve

Exhaust <4 t / * \ J

Vent Shut Off

Valve

Cryogenic Trap Cooler

(Liquid Argon)

Humidifier

Zero Shut Off

Valve

Manifold

Optional Isothermol

Oven

Figure 10. Canister c lean ing system.

Page 15-56 Compendium of Methods for Toxic Organic Air Pollutants January 1999

Method TO-15 VOCs

! m i «!

TIME

(b). Contaminated Sampler

Figure 12. Example of humid zero air test results for a clean sample canister (a) and a contaminated sample canister (b).

Page 15-58 Compendium of Methods for Toxic Organic Air Pollutants January 1999

3-WAY VALVE

Figure 14. Water method of standard preparation in canisters.

Page 15-60 Compendium of Methods for Toxic Organic Air Pollutants January 1999

STATUS:

TRAP 1: Sampling TRAP 2: Desorbing

CLOSED A oreNift Mrc PUMP MT£T

SAMPLE PUMP

SAMPLE INLET

CAL/INT STD VENT

PURGE GAS

CALGAS

INTERNAL STD

| MFC |-

• t><3 SV-2I

f—C*r—<»

sv

•A—<> A—

PUgGE GAS SAMPLE___PUnGFi VENT

IILLIUM

SOLID SORDENT CONCENTRATOR

i '>:'>:'>:'>:'>:'>:'>:'>:\

iiiiiiiiiiiiiiiiii 1

TO GC/ ; »»

DETECTOR

STIRLING CYCLE COOLER

Figure 16r Sample flow diagram of a commercially-available concentrator showing the combination of multisorbent ti.be and cooler (Trap 1 sampling; Trap 2 desorbing).

APPENDIX D

U.S. EPA Environmental Response Team (ERT)

Standard Operating Procedure (SOP) #1704: SUMMA Canister Sampling

July 1995

m U. S. EPA ENVIRONMENTAL RESPONSE TEAM W •

STANDARD OPERATING PROCEDURES SOP:

PAGE: REV:

DATE:

1704 1 of 15

0.1 07/27/95

SUMMA CANISTER SAMPLING

CONTENTS

1.0 SCOPE AND APPLICATION

2.0 METHOD SUMMARY

3.0 SAMPLE PRESERVATION, CONTAINERS, HANDLING, AND STORAGE

4.0 INTERFERENCES AND POTENTIAL PROBLEMS

5.0 EQUIPMENT/APPARATUS

5.1 Subatmospheric Pressure Sampling Equipment

5.2 Pressurized Sampling Equipment

6.0 REAGENTS

7.0 PROCEDURES

7.1 Subatmospheric Pressure Sampling 7.1.1 Sampling Using a Fixed Orifice, Capillary, or Adjustable Micrometering Valve ' 7.1.2 Sampling Using a Mass Flow Controller/Vacuum Pump Arrangement (Andersen Sampl

87-100)

7.2.1 Sampling Using a Mass Flow Controller/V acuum Pump Arrangement (Andersen Sampler Model 87-100)

7.2 Pressurized Sampling

8.0 CALCULATIONS

9.0 QUALITY ASSURANCE/QUALITY CONTROL

10.0 DATA VALIDATION

11.0 HEALTH AND SAFETY

U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP:

PAGE: REV:

DATE:

1704 3 of 15

•0.1 07/27/95

SUMMA CANISTER SAMPLING

1.0 SCOPE AND APPLICATION

The purpose ofthis standard operating procedure (SOP) is to describe a procedure for sampling of volatile organic compounds (VOCs) in ambient air. The method is based on samples collected as whole air samples in Summa passivated stainless steel canisters. The VOCs are subsequently separated by gas chromatography (GC) and measured by mass-selective detector or multidetector techniques. This method presents procedures for sampling into canisters at final pressures both above and below atmospheric pressure (respectively referred to as pressurized and subatmospheric pressure sampling).

This method is applicable to specific VOCs that have been tested and determined to be stable when stored in pressurized and subatmospheric pressure canisters. The organic compounds that have been successfully collected in pressurized canisters by this method are listed in the Volatile Organic Compound Data Sheet (Appendix A). These compounds have been measured at the parts per billion by volume (ppbv) level.

These are standard (i.e., typically applicable) operating procedures which may be varied or changed as required, dependent on site conditions, equipment limitations or limitations imposed by the procedure or other procedure limitations. In all instances, the ultimate procedures employed should be documented and associated with the final report.

Mention of trade names or commercial products does not constitute U.S. EPA endorsement or recommendation for use.

2.0 METHOD SUMMARY

Both subatmospheric pressure and pressurized sampling modes use an initially evacuated canister. Both modes may also use a mass flow controller/vacuum pump arrangement to regulate flow. With the above configuration, a sample of ambient air is drawn through a sampling train comprisedof components that regulate the rate and duration of sampling into a pre-evacuated Summa passivated canister. Alternatively, subatmospheric pressure sampling may be performed using a fixed orifice, capillary, or adjustable micrometering valve in lieu of the mass flow controller/vacuum pump arrangement for taking grab samples or short duration time-integrated samples. Usually, the alternative types of flow controllers are appropriate only in situations where screening samples are taken to assess for future sampling activities.

3.0 SAMPLE PRESERVATION, CONTAINERS, HANDLING, AND STORAGE

; After the air sample is collected, the canister valve is closed, an identification tag is attached to the canister, and the canister is transported to a laboratory for analysis. Upon receipt at the laboratory, the canister tag data is recorded. Sample holding times and expiration should be determined prior to initiating field activities.

4.0 INTERFERENCES AND POTENTIAL PROBLEMS

UBHBj U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP

PAGE REV

DATE SUMMA CANISTER SAMPLING

4. Particulate matter filter - 2- m sintered stainless steel in-line filter (Nupro Co., Model SS-2F-K4-2, or equivalent).

5. Cliromatographic grade stainless steel tubing and fittings - for interconnections (Alltech Associates, Cat. #8125, 'or equivalent). All materials in.contact with sample, analyte, and support gases should be chromatographic grade stainless steel.

6.0 REAGENTS

This section is not applicable to this SOP.

7.0 PROCEDURE

7.1 Subatmospheric Pressure Sampling

7.1.1 Sampling Using a Fixed Orifice, Capillary, or Adjustable Micrometering Valve

Prior to sample collection, the appropriate information is completed on the Canister Sampling Field Data Sheet (Appendix C).

A canister, which is evacuated to 0.05 mm Hg and fitted with a flow restricting device, is opened to the atmosphere containing the VOCs to be sampled.

The pressure differential causes the sample to flow into the canister.

This technique may be used to collect grab samples (duration of 10 to 30 seconds) or time-integrated samples (duration of 12 to 24 hours). The sampling duration depends on the degree to which the flow is restricted.

A critical orifice flow restrictor will have a decrease in the flow rate as the pressure approaches atmospheric.

Upon sample completion at the location, the appropriate information is recorded on the Canister Sampling Field Data Sheet.

7.1.2 Sampling Using a Mass Flow Controller/V acuum Pump Arrangement (Andersen Sampler Model 87-100)

1: Prior to sample collection the appropriate information is completed on the Canister Sampling Field Data Sheet (Appendix C).

1704 5 of 15

0.1 07/27/95

1.

2.

3.

4.

5.

6.

bRBM U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP: 1704

PAGE: 7 of 15 REV: 0.1

DATE: 07/27/95 SUMMA CANISTER SAMPLING

8.0 CALCULATIONS

1. A flow control device is chosen to maintain a constant flow into the canister over the desired sample period. This flow rate is determined so the canister is filled to about 88.1 kPa for subatmospheric pressure sampling or to about one atmosphere above ambient pressure for pressurized sampling over the desired sample period. The flow rate can be calculated by:

F WW (7X60)

where: •

flow rate (cmVmin) final canister pressure, atmospheres absolute volume of the canister (cm3) sample period (hours)

For example, if a 6-L canister is to be filled to 202 kPa (two atmospheres) absolute pressure in 24 hours, the flow rate can be calculated by:

F W 0 0 0 ) 8 3 c / n 3 / m i n

(24)(60)

F P V T

If the canister pressure is increased, a dilution factor (DF) is calculated and recorded on the sampling data sheet.

DF I* Xa

where:

U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP

PAGE REV

DATE

1704 9 of 15

0.1 07/27/95

SUMMA CANISTER SAMPLING

3. J. F. Walling, "The Utility of Distributed Air Volume Sets When Sampling Ambient Air Using Solid Adsorbents," Atmospheric Environ., 18:855-859, 1984.

4. J. F. Walling, J. E. Bumgarner, J. D. Driscoll,C. M. Morris, A. E. Riley, and L. H. Wright, "Apparent Reaction Products Desorbed From Tenax Used to Sample Ambient Air," Atmospheric Environ., 20:51 -57, 1986.

5. Portable Instruments User's Manual for Monitoring VOC Sources, EPA-340/1 -88-015, U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Washington, D.C., June 1986.

6. R. A. Rasmussen and J. E. Lovelock, Atmospheric Measurements Using Canister Technology, J. Geophys. Res., 83: 8369-8378, 1983.

R. A. Rasmussen andM. A. K. Khalil, "Atmospheric Halocarbon: Measurements and Analysis of Selected Trace Gases," Proc. NATO ASI on Atmospheric Ozone, BO: 209-231.

EPA Method TO-14 "Detennination of Volatile Organic Compounds (VOCs) in Ambient Air Using Summa Passivated Canister Sampling and Gas Chromatographic Analysis", May 1988.

APPENDIX A Volatile Organic CompoundiData Sheet

SOP #1704 July 1995

U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP

PAGE REV

DATE SUMMA CANISTER SAMPLING

APPENDIX B Figure

SOP #1704 July 1995

1704 11 of 15

0.1 07/27/95

mi U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP

PAGE REV

DATE

1704 13 of 15

0.1 07/27/95

SUMMA CANISTER SAMPLING

APPENDIX C Canister Sampling Field Data Sheet

SOP #1704 July 1995

APPENDIX E

U.S. EPA Environmental Response Team (ERT)

Standard Operating Procedure (SOP) #170: Sample Documentation

September 1994

| | | p U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES . ' SOP: 1703

,; PAGE: l o f l O REV: 0.0

•. DATE: 09/01/94 SAMPLE DOCUMENTATION

:CONTENTS

I . 0 SCOPE AND APPLICATION ''

2.0 METHOD SUMMARY

3.0 SAMPLE PRESERVATION, CONTAINERS, HANDLING AND STORAGE

3.1 Canister Receipt

3.2 Canister Storage

4.0 INTERFERENCE AND POTENTIAL PROBLEMS '

5.0 EQUIPMENT/APPARATUS

5.1 Canister .; '

5.2 Canister Cleaning System

6.0 REAGENTS ' ,

7.0 PROCEDURE

7.1 System Set-Up 7.2 Cleaning

7.3 Leak-Testing . ' ;•• . [

8.0 CALCULATIONS

9.0 QUALITY ASSURANCE/QUALITY CONTROL ! .

10.0 DATA VALIDATION ' '

I I . 0 HEALTH AND SAFETY

12.0 REFERENCES : :

13.0 APPENDIX

A - Figures

SUPERCEDES: SOP #1703; Revision 2.1; 05/24/91; U.S. EPA Contract 68-03-3482.

U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP:

PAGE: REV:

DATE:

1703 2 of 10

0.0 09/01/94

SAMPLE DOCUMENTATION

1.0 SCOPE AND APPLICATION

This standard operating procedure (SOP) is intended for use when cleaning Summa polished stainless steel canisters. Summa canisters provide a medium to sample gas-phase Volatile Organic Compounds (VOCs) on-site at concentrations of one part per billion by volume (ppbv) and greater. This procedure is to assure that canisters have been sufficiently cleaned prior to sampling, to the extent that no VOC contamination is present at concentrations greater than 0.2 ppbv.

These are standard (i.e., typically applicable) operating procedures which may be varied or changed as required, dependent on site conditions, equipment limitations or limitations imposed by the procedure or other procedure limitations. In all instances, the ultimate procedures employed should be documented and associated widi the final report.

Mention of trade names or commercial products does not constitute U.S. EPA endorsement or recommendation for

2.0 METHOD SUMMARY

After use, canisters are logged in and physically inspected. These canisters are vented to the outside air under an operating exhaust hood. Canisters are connected to a manifold which is attached to a vacuum pump via a cryogenic trap. The canisters and lines are evacuated and then the canisters are heated to an elevated temperature for a prescribed time period. During the heating period, the canisters are filled with humidified nitrogen and pressurized. The process is repeated. The filling and pressurizing functions are followed by evacuation and heating and are performed a total of three times.

Canisters are confirmed free of VOC contamination by pressurizing the canisters with ultra high purity nitrogen and analyzing on the GC/MS. If no VOC contamination is present at concentrations greater than 0.2 ppbv, the canister is determined clean. Clean canisters are leak-tested by pressurizing with nitrogen for 24 hours. Canisters that have

. been determined clean and without leaks are evacuated. These canisters are logged in as cleaned and certified and are stored in the evacuated state with brass cap fittings until needed for sampling.

3.0 SAMPLE PRESERVATION, CONTAINERS, HANDLING AND STORAGE

3.1 Canister Receipt

use.

The overall condition of each sample canister is observed. Any canister having physical defects requires corrective action.

2 Each canister should be observed for an attached sample identification number.

Each canister is recorded in the dedicated laboratory logbook by its Summa canister number.

3.2 Canister Storage

« g f | ) U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP

PAGE REV

DATE SAMPLE DOCUMENTATION

6.0 REAGENTS

Gas cylinders of nitrogen, ultra high purity grade.

Cryogen - liquid nitrogen (bp -195°C).

Distilled, deionized water, ultra high purity.

7.0 PROCEDURE •

7.1 System Set-Up

1. All connections in the vacuum system except the canisters and manifold are sealed. All connections, lines, and valves are checked for leaks by pressurizing the line to 30 psig and using a soap solution. The septum is checked for leaks by visual inspection.

2. The liquid nitrogen is added to the cryogenic trap and allowed to equilibrate.

3. Check the pump to assure proper working order by achieving a vacuum of 0.05 mm Hg in the line that normally attaches to the manifold but is now capped. Valve A is open and Valve B is closed. After the vacuum test is completed, turn the pump off and remove the cap to break the vacuum.

4. Check the oven to assure proper working order by heating the oven to 100°C and measuring the internal temperature with a thermometer.

5. Check reagents to assure proper purity.

6. Set the back pressure on the nitrogen to 30 psig.

7.2 Cleaning

1. All canisters are vented to the outside air under an operating exhaust hood.

2. Connect the canisters (with the valves closed on the canisters) to the manifold by the Swagelok fittings. Connect the manifold to the vacuum system by the Swagelok fitting.

3. Open Valve A, assure Valve B is closed, and start vacuum pump.

4. Once a vacuum (0.05 mmHg) is obtained in the line and the manifold, Valve A is closed. The system is then examined for leaks by comparing the initial vacuum reading and a second vacuum reading three minutes later. I f the vacuum deteriorates more than 5 mm Hg, a leak exists and corrective action, such as tightening all fittings, is necessary.

1703 4 of 10

0.0 09/01/94

5. If no leaks are observed, Valve A is opened and the Canister 1 valve is opened. Evacuate Canister 1 to 0.05 mm Hg, then close Canister 1 valve. By evacuating one canister at a time, cross contamination between canisters is minimized.

y U, S. EPA ENVIRONMENTAL RESPONSE TEAM A;

STANDARD OPERATING PROCEDURES SOP:

PAGE: REV:

DATE:

1703 6 of 10

0.0 09/01/94

SAMPLE DOCUMENTATION

7.3 Leak-Testing

Once the canister lot is determined as being clean, the canisters are pressurized to 30 psig with nitrogen..

2 The initial pressure is measured via the pressure gauge, die canister valve is closed, and die brass cap is replaced. Document the time and pressure.

After 24 hours, the final pressure is checked. Document the time and pressure.

4. If leak tight, the pressure should not vary more than ±13.8 kPa (±2 psig) over the 24-hour period. If this criterion is met, the canister is capped with a brass fitting and stored. If a leak is present, corrective action such as tightening all fittings, is required. Document the results.

8.0 CALCULATIONS

There are no calculations for this SOP.

9.0 QUALITY ASSURANCE/QUALITY CONTROL -

The following specific quality assurance/quality control procedures are applicable for Summa canister cleaning

1. • All connections, lines, and valves are checked to assure no leaks are present.

2. The septum is checked, to assure no leaks are present, by removing the septum and visually examining it.

3. The pump is checked to assure proper working order by achieving a vacuum of 0.05 mm Hg prior to cleaning.

4. The oven is checked to assure proper working order by comparing the oven setting at 100°C to the internal temperature with a thermometer.

5. The reagents are checked to assure sufficient purity.

6. All canisters are to be evacuated to 0.05 mmHg during each cycle of the cleaning process and the results are to be documented. '

7. All canisters are to be evacuated at 100°C for one hour during each cycle of the cleaning process. Results are to be documented.

8. All canisters are to be evacuated, heated, and pressurized three times during the cleaning process. Document each cycle.

9. The selected canister from the cleaning lot to be tested must be analyzed by GC/MS as shown to be sufficiently cleaned to the extent that no VOC contamination is present at concentrations greater than 0.2 ppbv for the canister lot to be considered cleaned. If the VOC contamination is greater than 0.2 ppbv, the canister lot must be cleaned again. In either case, the results will be documented.

(iKHRJ U. S. EPA ENVIRONMENTAL RESPONSE TEAM

STANDARD OPERATING PROCEDURES SOP: 1703

PAGE: 8 of 10 REV: 0.0

DATE: 09/01/94 SAMPLE DOCUMENTATION

APPENDIX A Figures

SOP #1703 September 1994

APPENDIX F

ERG's Method Detection List

Compound ppbv | Compound ppbv [Ethyl Aery late 0.16 | Bromodichloromethane 0.06 jTrichloroethylene 0.07 | Methyl Methacrylate 0.18 \cis-1,3-Dichloropropene 0.1 |Methyl Isobutyl Ketone 0.15 \trans-1,3-Dichlordpropene 0.11 |1,1,2-Trichloroethane 0.06 |Toluene 0.06 |Dibromochloromethane 0.08 |1,2-Dibromoethane 0.08 ui-Octane 0.06 |Tetrachlproethylene 0.06 jChlorobenzene 0.06 jEthylbenzene 0.04 ]m,p-Xylene 0.05 |Bromoform 0.08 |Styrene 0.07 |1,1,2,2-Tetrachloroethane 0.06 |0-Xy/ene 0.05 j 1,3,5-Trimethylbenzene 0.07 11,2,4-Trimethylbenzene 0.07 | m-Dichlorobenzene 0.05 | Chloromethylbenzene 0.07 \p-Dichlorobenzene 0.09 \o-Dichlorobenzene 0.06 11,2,4-Trichlorobenzene 0.06 |Hexachloro-1,3-Butadiene 0.06

[Acetylene 0.13 [Propylene 0.05 jDichlorodifluoromethane 0.04 IChloromethane 0.06 iDichlorotetrafluoroethane 0.05 iVinyl Chloride 0.06 :1,3-Butadiene 0.07 iBromomethane 0.09 Chloroethane 0.08 Acetonitrile 0.25 [Acetone 0.26 iTrichlorofluoromethane 0.04 iAcrylonitrile 0.21 h,1-Dichloroethene 0.1 ! Methylene Chloride 0.06 jTrichlorotrifluoroethane 0.07 \ trans-1,2-Dichloroethylene 0.06 1,1,-Dichloroethane 0.08 Methyl tert-Butyl Ether 0.18 Methyl Ethyl Ketone 0.15 Chloroprene 0.1 cis-1,2-Dichloroethene 0.1 Bromochloromethane 0.12 Chloroform 0.05 Ethyl tert-Butyl Ether 0.15 1,2-Dichloroethane 0.06 1,1,1-Trichloroethane 0.06 Benzene 0.04 Carbon Tetrachloride 0.08 tert-Amyl Methyl Ether 0.12 1,2-Dichloropropane 0.07

APPENDIX G

Example Questionnaire

Example Canister Field Data Sheet

and

Example Chain of Custody

Project No. Element No. Revision No. Date Page

0121.00 B2

1 March 2000.

2 of 17

IBS EASTERN RESEARCH GROUP, INC

Canister Sample Data Sheet LAB ID#

UJ u.

CD z -> o LU DC.:

o O

Location: City / State Sampling Period: Nox Analyzer Operating: Average PPM:

. Elapsed Time:

Average Wind Speed: _ Average Wind Direction: Average Temperature: Average Barometric Pressure: Relative Humidity: Flow Controller Set at: Comments:

Site Code: Collection Date:

Canister Number: Operator:

Initial Vacuum: Final Field Pressure/Vacuum:

Duplicate (YIN) Duplicate Can #:

Options Flow Controller Zero Reading

Received by: Date Received: Carbonyl Tubes:

Carbonyl ID #: _

Pressure @ Receipt: Void Acceptable Yes No

Stored:

Analyst: Analysis Date: Analysis Time: NMOC Instrument:

Area Counts run 1: ppmC run 1:

Canister Number: Analysis Pressure: Sample Replicate:

Initial or Repeat:

Average AC: Standard Dev:

Entered into Database by:

Area Counts run 2: ppmC run 2: •

Average: ppmC: Standard Dev:

Area Counts run 3: ppmC run 3:

Date:

o o 5 z w

Analyst: _ _ _ _ _ Analysis Pressure: Load Volume:

Date: Data File Name: Duplicate File Name:

Date:

g " X "

o I V

Analyst: Analysis Pressure: Load Volume:

Date: Data File Name: Duplicate File Name: Replicate File Name:

White: Sample File Copy Yellow: Receiving Copy

Figure 7-1. Canister Sample Data Sheet

Pink: Field Copy

glp/D:\SECT7,WPD

Site:

Date:

I.D. #

Channel PPMC IB:

ra­ta

Figure 7-3. Canister Tag

US EPA REGION 2

CHAIN OF CUSTODY/ FIELD DATA FORM Page of pages

SURVEY NAME & LOCALITY

PROGRAM: SF'("~

Permit #

SITE ID OPERABLE UNIT PROJECT LEADER

PROGRAM RESULTS CODE

RCRA • NPDES • SDWA • AM _ CAA • TSCA |~1 ENFORCEMENT: CRIMINAL O CIVIL •

LAB ID/ FIELD ID

o Z £ SPECIAL ni « =H REQUIRE-

DESCRIPTION & INSTRUCTIONS INCLUDING LOCATION, ESTIMATED CONCENTRATIONS, SPECIAL REPORTING LIMITS, SPECIAL TEST REQUIREMENTS & ALIQUOTING

Preservative

(circle)

Collection Time (24hr clock) //////////////

Collection Date

u 1 2 3 4 5 6 7 8 E

11 I I i i / u u / y y

• 1 2 34 5 6 7 8S

L I 1 2 3 4 5 6 7 89

• .1 2 3 4 5 6 7 8 9

• 1 2 3 4 5 6 7 8 9

• 12 3 4 5 6 7 8 9

I I • ' • 1 2 3 4 5 6 7 8 9

• • • 12 3 4 5 6 7 8 9

u 1 2 3 4 5 6 7 8 9

U 1 2 3 4 5 6 7 8 9

C O M M b N I S : •

Matrix: A=aqueous B=aqueous (chlorinated) C=soil D=sediment E=sludge

F=multiphasic G=solvent H=biota l=oil J=other

Survey Complete? Y • N •

Relinquished By:

Relinquished By:

Relinquished By:

1=ice 2=H2S04 pH<2 3=HN03 pH<2 4=HCI pH<2 5=Na2S203 6=NaOH pH>9 7=Ascorbic Acid

8 = FAS 9=ZnAc

Time

Person Assuming Responsibility for Sample(s):

Received By:

Received By:

Received By:

Date

INDOOR AIR QUALITY- BUILDING SURVEY

Occupant/Building Name: Date:

Address:

Completed by: Case #

Building type: residential/office/commercial/industrial

Basement size: ft3

Number of floors

below grade: - (full basement/crawl space/slab)

at or above grade:

Foundation type: poured concrete (over gravel) /cinder blocks/earthen/ other (specify)

Buildings occupants: Children under age 13 Children age 13-18 Adults

Contaminant Source Category Yes No Comments/Locations

OUTSIDE SOURCES

Garbage dumpsters

Heavy motor vehicle traffic

Construction activities

Nearby industries (identify)

UST/AST (gasoline, heating fuel)

BASEMENT SURVEY

Wall construction (cinder block, sheet rock, paneling, etc.)

type: condition:

Floor Construction (earthen, slab, floating, etc)

type: condition:

Number of windows present on each wall and size North: East: South: West:

Was basement painted recently? oil-base or latex paints

date: type of paint:

Sump present (PID/FID/CGI#s?)

YES NO

Location of sump

New flooring in basement?(list type - carpet, tile)

using glue

New furniture added to basement type: date:

Staining on floors/walls

Moisture visually present in the basement

Pipes running through walls, floor (conduits-describe, give FID/PID/CGS readings)

Odors detected by inspector

Basement used as living space

Time occupants spend in basement (hours/day/per person)

Items stored in basement: solvents

gasoline

paint/thinners

polishes/waxes

insecticides

kerosene

household cleaning products

mothballs

other items?

NOTES:

FIRST FLOOR SURVEY

Wall construction (cinder block, sheet rock, paneling, etc)

> type: condition:

Was painting done recently? oil-base or latex?

date: type of paint:

New flooring on 1 s t floor? (list type - carpet, tile)

using glue

YES NO

New furniture added to 1 s t floor? (list type - carpet, tile)

type: date:

Staining on floors/walls

Pipes running through walls, floor (describe)

Odors detected by inspector

Items stored on this floor

solvents

gasoline

paint/thinners

polishes/waxes

, insecticides

kerosene

household cleaning products

mothballs

other items?

NOTES:

PERSONAL ACTIVITIES

Does anyone in the building smoke?

approx. number of tobacco products per day, per person

Does anyone dry-clean their clothes?

List hobbies of Residents

Any house pets?

MISCELLANEOUS

Have the occupants ever noticed unusual odors in building ?

describe: location:

Known spill outside or inside building (Specify location)

Type of heating used in building oil

natural gas

kerosene

YES NO

electric

other (specify)

If heating oil, identify the location and age of the storage tank

Is the heating unit properly vented?

Water damage or standing water in building (historic or current)

Fire damage to building date:

Pest control applications date:

Septic system

FIELD SCREENING RESULTS FID pro CGI C02 Rel. Hum

Basement

First Floor

Additional Floors

Other (specify)

APPENDIX H

Resident Instructions

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

REGION II

The United States Environmental Protection Agency is conducting an investigation of air quality in residential homes in your area. Concerns over contamination allegedly linked to a former dry cleaning operation has sparked local and federal attention. As result, the US EPA will be corning to your home to discuss this matter with you, and collect air samples from your basement.

PUBLIC FACT SHEET r AIR SAMPLING

Why is the United States Environmental Protection Agency collecting air samples from my basement?

An agreement has been formulated by local, state, and federal environmental agencies that your home should be tested for indoor air quality. As a result, US EPA personnel will conduct the sampling of air from your basement, at no charge to you. The sampling of air is to deterrnine whether you and/or your family are at risk of breathing harmful contaminants that may be associated with local environmental issues.

How is the air being collectedfrom my basement?

A device, known as a SUMMA Canister, will be placed in your basement to draw in air for a period of 24 hours. Initially, pressure inside the canister is set at a lower pressure than that of the air in your basement. During the 24 hour time frame, air will flow into the SUMMA canister until the pressure of air inside equals the pressure outside the canister. Air will not flow out of the device. These canisters are completely safe and pose no danger to you or your children.

Who is doing the analysis of the samples?

While US EPA personnel are collecting the air samples, a private laboratory has been contracted to perform the analytical procedures. .

What should I NOT do so that I do not damage or disrupt the sampling device?

Air sampling devices are particularly sensitive, and can be damaged very easily. This is why it is important to practice the following precautions, starting 24 hours prior to sampling:

-do not smoke in the basement -minimize your movement around the device -do not bring dry-cleaning in the house -do not use solvents of any type -do not open your basement windows -do not utilize fans or vents in the basement -do not paint or clean paint brushes . -do not polish your shoes " -do not pour gasoline or liquid fuels inside your house or attached garage -do not move the canister(s) under any circumstances.

The US EPA apologizes for any inconveniences that may occur as part of this sampling event. However, your cooperation and understanding is greatly appreciated. Remember, we are doing this for the protection of your health, as well as the surrounding community's.

EPA Contact: Andrew Confortini 732/906-6827 confortini.andrew@epa.gov