HOT MIX ASPHALT PLANTS - EPA...Hot Mix Asphalt Plants Technical Systems Audit of Testing at Asphalt...

United States Office Of Air Quality EPA-454/R-00-026Environmental Protection Planning And Standards May 2000Agency Research Triangle Park, NC 27711

Air

HOT MIX ASPHALT PLANTS

TECHNICAL SYSTEMS AUDIT ofTESTING at PLANT C

ASPHALT PLANT CLOS ANGELES, CALIFORNIA

This

is

a

Blank

Page

EPA-454/R-00-026

HOT MIX ASPHALT PLANTSTECHNICAL SYSTEMS AUDIT of TESTING at PLANT C

ASPHALT PLANT CLOS ANGELES, CALIFORNIA

U.S. ENVIRONMENTAL PROTECTION AGENCYOffice of Air and Radiation

Office of Air Quality Planning and StandardsResearch Triangle Park, North Carolina 27711

May 2000

This

is

a

Blank

Page

ii

iii

DISCLAIMER

The information in this document has been funded wholly or in part by the Office of AirQuality Planning and Standards, U.S. Environmental Protection Agency (EPA) under contract68-D4-0091 to Research Triangle Institute. It has been reviewed and revised by the EmissionsMeasurement Center Quality Assurance Officer. It has been approved for publication as an EPAdocument. Mention of trade names or commercial products is not intended to constituteendorsement or recommendation for use.

This

is

a

Blank

Page

iv

v

Preface

This report was produced by the Source Measurement Technology Group of EPA’s EmissionsMeasurement Center located in Research Triangle Park, NC. It is one of a series of twelvereports prepared to document an EPA emission test program to characterize emissions to the airfrom hot mix asphalt plants. These twelve reports and their associated EPA document numbersand publication dates are:

Document TitleEPA Document

NumberPublication

Date

Hot Mix Asphalt PlantsEmission Assessment Report EPA 454/R-00-019

Hot Mix Asphalt PlantsKiln Dryer Stack Instrumental Methods TestingAsphalt Plant A, Cary, North Carolina EPA 454/R-00-020 April 2000

Hot Mix Asphalt PlantsKiln Dryer Stack Manual Methods TestingAsphalt Plant A, Cary, North Carolina

Volume 1 of 2 EPA 454/R-00-021a April 2000

Volume 2 of 2 EPA 454/R-00-021b April 2000

Hot Mix Asphalt PlantsKiln Dryer Stack Instrumental Methods TestingAsphalt Plant B, Clayton, North Carolina EPA 454/R-00-022 April 2000

Hot Mix Asphalt PlantsKiln Dryer Stack Manual Methods TestingAsphalt Plant B, Clayton, North Carolina

Volume 1 of 2 EPA 454/R-00-023a April 2000

Volume 2 of 2 EPA 454/R-00-023b April 2000

Hot Mix Asphalt PlantsTruck Loading and Silo Filling Instrumental Methods TestingAsphalt Plant C, Los Angeles, California

EPA 454/R-00-024 May 2000

Hot Mix Asphalt PlantsTruck Loading and Silo Filling Manual Methods TestingAsphalt Plant C, Los Angeles, California

Volume 1 of 8 EPA 454/R-00-025a May 2000

Volume 2 of 8 EPA 454/R-00-025b May 2000

Volume 3 of 8 EPA 454/R-00-025c May 2000

Volume 4 of 8 EPA 454/R-00-025d May 2000

Volume 5 of 8 EPA 454/R-00-025e May 2000

Volume 6 of 8 EPA 454/R-00-025f May 2000

Volume 7 of 8 EPA 454/R-00-025g May 2000

Volume 8 of 8 EPA 454/R-00-025h May 2000

Document TitleEPA Document

NumberPublication

Date

vi

Hot Mix Asphalt PlantsTechnical Systems Audit of Testing at Asphalt Plant CAsphalt Plant C, Los Angeles, California EPA 454/R-00-026 May 2000

Hot Mix Asphalt PlantsTruck Loading Instrumental Methods TestingAsphalt Plant D, Barre, Massachusetts EPA 454/R-00-027 May 2000

Hot Mix Asphalt PlantsTruck Loading Manual Methods TestingAsphalt Plant D, Barre, Massachusetts EPA 454/R-00-028 May 2000

Hot Mix Asphalt PlantsResponse to Comments on Testing Program for AsphaltPlants C and D EPA 454/R-00-029 May 2000

Hot Mix Asphalt PlantsStakeholders Opinions Report EPA 454/R-00-030

These documents, including this Response to Comments document, are available for downloading, onCD-ROM and in paper.

Downloads can be made from:

http//www.epa.gov/ttn/emc/asphalt.html

Copies of the CD ROM can be requested by mail at:

Emission Measurement Center, MD-19US Environmental Protection AgencyResearch Triangle Park, NC 27711

Paper copies of the reports can be obtained from:

National Technical Information Service5285 Port Royal RoadSpringfield, VA 22161Phone orders 1-800-553-6847 or (703) 605-6000; FAX orders (703) 605-6900http://www.ntis.gov/products/environment.htm

Acknowledgments

Many individuals contributed to the development of this report. Laura Autry of EPA’s Air QualityTrends Analysis Group, R.K.M. Jayanty of Research Triangle Institute (RTI) and Robert S. Wright ofRTI are the primary authors of the report. David Mobley, Acting Director of EPA’s EmissionsMonitoring and Analysis Division, Conniesue Oldham of EPA’s Air Quality Trends Analysis Group,Bill Lamason and Gary McAlister of EPA’s Emission Measurement Center provided advice inresolving some findings identified by the RTI audit team.

vii

Glossary

ASTM – American Society for Testing MaterialsCO – Carbon MonoxideCTS – Calibration Transfer StandardEPA – United States Environmental Protection AgencyFID – Flame Ionization DetectorFTIR – Fourier Transform Infrared SpectroscopyHAP – Hazardous Air PollutantMCEM – Methylene Chloride Extractable Particulate Matter or Organic Extractable Particulate

MatterMQL – Minimum Quantitation LimitMRI – Midwest Research InstituteNIST – National Institute of Standards and TechnologyNOx – Nitrogen OxidesOAQPS – Office of Air Quality Planning and StandardsPAH – Polynuclear Aromatic HydrocarbonsPES – Pacific Environmental Services, Inc.PM – Particulate Matterppb – parts per billionppm – parts per millionQAPP – Quality Assurance Project PlanRAP – Recycled Asphalt PavementRSD – Relative Standard DeviationsRTI – Research Triangle InstituteSED – Silo Exhaust DuctSF6 – Sulfur HexafluorideSO2 – Sulfur DioxideSSTP – Site Specific Test PlanSVOHAP – Semi-volatile Organic Hazardous Air PollutantTHC – Total HydrocarbonsTSA – Technical Systems AuditVOHAP – Volatile Organic Hazardous Air PollutantVOST – Volatile Organic Sampling Train

This

is

a

Blank

Page

viii

ix

Table of Contents

Disclaimer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Chapter 1

1.0 Quality Assurance Audit Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.11.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.11.2 Process Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1

1.2.1 Aggregate Processing Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.21.2.2 Silo Filling Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.21.2.3 Load-Out Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2

1.3 Emission Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.21.4 Technical Systems Audit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3

1.4.1 General Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.41.4.2 Technical Assessment of MRI’s Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.41.4.3 Technical Assessment of PES’s Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5

Chapter 2

2.0 Technical Systems Audit - Midwest Research Institute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1Technical Systems Audit Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4

A. Quality System Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4B. Organization and Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7C. Training and Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.10D. Data Quality Indicator Goals and Performance Testing . . . . . . . . . . . . . . . . . . . . . . . . . 2.10E. Extractive Sampling System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.12F. Extractive Fourier Transform Infrared (FTIR) Instrument . . . . . . . . . . . . . . . . . . . . . . . 2.16G. Sample Concentrator for Ftir Instrument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.24H. Total Hydrocarbons (THC) Analyzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.32I. Capture Efficiency Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.37

Chapter 3

3.0 Technical Systems Audit - Pacific Environmental Services Inc. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1Technical Systems Audit Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2

A. General Quality Assurance Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2B. Organization and Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5C. Method Specific - EPA Method 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7

x

D. Method Specific - EPA Method 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8E. Method Specific - EPA Method 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.9F. Method Specific - EPA Method 315 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.10

Laboratory Information Necessary for EPA Method 315 . . . . . . . . . . . . . . . . . . . . . 3.12Calibration and General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.13

G. Method Specific - EPA Method 18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.16Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.17

H. Method Specific - SW-846 Method 0030 (VOST) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.19I. Method Specific - SW-846-0010 (Modified Method 5) . . . . . . . . . . . . . . . . . . . . . . . . . 3.21

Train Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.22Preparation of the Train . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.23Running the Train . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.25Sample Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.25

J. Sampling General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.27K. Quality Assurance/quality Control/general . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.28L. Sample Custody and Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.29M. Miscellaneous Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.30N. Estimation of Particulate Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.30P. Estimation of Particulate Deposition on the Inside Walls of the Temporary

Silo Exhaust and Wind Tunnel Exhaust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.31Q. Direct Interface Portable GC/MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.32

Chapter 1

Quality Assurance Audit ResultsHot Mix Asphalt Plant C,Los Angeles, CA

This

is

a

Blank

Page

1.1Quality Assurance Audit Results

1.0 QUALITY ASSURANCE AUDIT RESULTS

1.1 Overview

The United States Environmental Protection Agency (EPA) Office of Air Quality Planning andStandards (OAQPS) is investigating hot mix asphalt plants to characterize emissions during silo fillingand truck loading operations. In support of this investigation, OAQPS issued work assignments toPacific Environmental Services, Inc. (PES) and Midwest Research Institute (MRI) to conduct emissiontesting of these sources. Additionally, due to the complexity and importance of this testing, OAQPSissued a work assignment to Research Triangle Institute (RTI) to perform an independent technical auditof MRI and PES’s emissions testing through a technical review. RTI conducted the audit according tothe principles of the EPA Office of Research and Development’s Quality Assurance Division. Theseprinciples are described in their working draft of EPA Guidance for Technical Assessments forEnvironmental Data Operations (EPA QA/G-7).

The primary objective of the emissions testing was to characterize the uncontrolled emissions ofparticulate matter (PM), organic extractable particulate matter (methylene chloride extractable particulateor MCEM) and organic hazardous air pollutants (HAP’s) from silo filling and truck loading operations ata hot mix asphalt production plant. The organic HAP emissions quantified included polynucleararomatic hydrocarbons (PAHs), semi-volatile organic hazardous air pollutants (SVOHAPs), and volatileorganic hazardous air pollutants (VOHAPs). Other emissions that were characterized included methane,carbon monoxide (CO), sulfur dioxide (SO2), and nitrogen oxides (NOx).

It was concluded from the technical systems audit that the overall quality assurance objectives ofthe test were met. Except for some minor deviations, the test team members performed the testingaccording to the procedures outlined in the Site Specific Test Plan (SSTP) and Quality Assurance ProjectPlan (QAPP). These deviations from the QAPP or SSTP were discussed with and approved by the EPAWork Assignment Manager. The equipment used was appropriate for the emissions testing and wasoperated satisfactorily during the testing.

1.2 Process Description

An asphalt plant located south of Los Angeles, CA, (also called Asphalt Plant C or Plant C), wasselected as the host facility. Testing was performed over five consecutive days beginning on July 24,1998. Testing of transport truck loading operations was performed under two conditions - normaloperations and a background condition. Additionally, a secondary objective of the emission testing wasto characterize controlled emissions of VOHAPs, CO, SO2 and NOx from the aggregate dryer.

This plant was selected for the emissions testing due to its high production rate, ventilation of thestorage silos and enclosure/ventilation of the load-out bay. The Plant C facility has a rated productioncapacity of 650 tons per hour (tph). Daily production varies from approximately 2,000 tons per day (tpd)to 6,000 tpd depending on demand. The plant produces five different categories of hot mix asphalt: 3/8in, 1/2 in, 3/4 in, fines, and recycled asphalt (RAP). These categories indicate the average size and typeof aggregate in the mix. In RAP, small amounts of recycled asphalt are added to the mix. The plant alsoadds small amounts of rubber to some products as a crack inhibitor. The plant uses two different kinds ofliquid asphalt: AR-4000 and AR-8000. AR-4000 is a softer asphalt and is used approximately 90% of


the time. The percent by weight of liquid asphalt in the mix varies from 4.8% to 6.0% depending on thesize of the aggregate (the smaller the aggregate, the higher the liquid asphalt content).

The following paragraphs describe the three operations tested at Plant C.

1.2.1 Aggregate Processing Operations

In this continuous process, cold aggregate is introduced to the rotary drum dryer. As the drumrotates, the aggregates move toward the other end of the drum. The cold aggregate is first dried and thenis heated as it moves through the drum. After exiting the dryer drum into a mixing drum, the heated anddried aggregate is mixed with the liquid asphalt cement and recycled asphalt pavement (RAP). Aventilation system exhausts the gases and condensed particulate from the rotary drum dryer through abaghouse and exhaust stack.

1.2.2 Silo Filling Operations

Hot mix asphalt produced in the aggregate processing operation is transported by bucket elevatorinto temporary storage silos. Plant C has five 200-ton heated silos that sit on top of the load-out tunnel. The silos serve as a holding station between production and the loading of the hot mix asphalt intotransport trucks. Depending on customer requirements, a different product can be stored in each silo. The hot mix asphalt in storage can have a temperature up to 160E C (320E F). A 10" inside diameterventilation duct exhausts the gases and condensed particulate from each of the five silos to an exhaustduct that ventilates the load-out tunnel.

1.2.3 Load-Out Operations

The hot mix asphalt is dropped from the storage silos into transport trucks within a load-outtunnel that is approximately 183 ft long with open doorways at both ends. During a full load-outschedule, trucks enter the tunnel approximately every one to three minutes. Single bed trucks holdapproximately 21 tons of asphalt cement. Dual bed trucks (i.e., a truck and trailer) hold approximately25 tons. Typically, the temperature of the asphalt cement, after it drops from the silo into the truck, isapproximately 300E F.

The truck is positioned under the silo containing the desired product where it is loaded into thetruck bed. During loading, emissions are captured by activating a double-slotted capture hood at eachsilo. With the truck positioned under the silo, one freestanding slot will be at the forward edge and oneat the aft edge of the truck bed. No more than one silo can operate at a given time and only the capturehood associated with that silo is activated to capture the emissions. It typically takes 15 to 30 seconds toload a truck. One capture hood is always active, even when no loading is occurring. Each of the capturehoods connects to the 32 inch inside diameter tunnel exhaust duct. Constant flow is maintained by thefan setting, thus, a constant airflow is always exhausted from the load-out tunnel to the emissionabatement system. The tunnel, ventilation system, and capture hoods work together to form a near-totalenclosure for the loading operations. The selected test site, however, did not meet all of the criteria for apermanent total enclosure as defined by EPA Method 204, “Criteria for Verification of a Permanent orTemporary Total Enclosure,” Federal Register, Vol. 62, No. 115, June 16, 1997.


1.3 Emission Testing

The emissions testing consisted of triplicate runs of the silo and load-out ventilation systems. Additionally, a fourth test run was performed using two trucks that traversed the load-out area likenormal operations, while no loading was occurring to determine background emissions contributed bydiesel truck exhaust. For all four test runs, capture efficiency testing of the load-out system wasperformed. The silo ventilation system was tested intermittently whenever silo loading operationsoccurred. Two test runs were performed on the dryer stack using only instrumental test methods. Thethree ventilation systems are referred to as the load-out system, the silo storage system, and the hot mixdryer system. The specific tests performed at each ventilation system are summarized below:

! The Load-out system was tested for HAPs, CO, SO2, and NOx, using extractive FourierTransform Infrared Spectroscopy (FTIR) (EPA Method 320) and FTIR with sampleconcentration. Total hydrocarbons (THC) were measured using a flame ionization detector (FID)(Method 25A). A single SW-846 Method 0010 Modified Method 5 (MM5) sampling train wasused to collect both PAHs and SVOHAPs. Three different procedures were used to measureVOHAPs: 1) sampling procedure SW 846 Method 0030 in combination with analyticalprocedure SW 846 Method 8260, 2) EPA Method 18, and 3) on-site GC/MS. During backgroundtesting no measurements were made using the on-site GC/MS.

! The Silo storage system was tested for HAPs, CO, SO2, and NOx, using extractive FTIR (EPAMethod 320) and FTIR with sample concentration. Total hydrocarbons (THC) were measuredusing a flame ionization detector (FID) (Method 25A). A single SW-846 Method 0010 ModifiedMethod 5 (MM5) sampling train was used to collect both PAHs and SVOHAPs. Three differentprocedures were used to measure VOHAPs: 1) sampling procedure SW 846 Method 0030 incombination with analytical procedure SW 846 Method 8260, 2) EPA Method 18, and 3) on-siteGC/MS.

! The Hot mix dryer system was tested for HAPs, CO, SO2, and NOx, using extractive FTIR (EPAMethod 320) and FTIR with sample concentration. Total hydrocarbons (THC) were measuredusing a flame ionization detector (FID) (Method 25A).

! Capture efficiency tests of the load-out system were also performed simultaneously with theload-out system tests. Tracer gas was released from a manifold in the load-out bay, was collectedby the ventilation system, and air concentrations were measured, allowing capture efficiency tobe calculated.

The responsibilities for this testing were divided between PES and MRI. The instrumental testmethods were performed by MRI under the direction of Scott Klamm. The manual test methods wereperformed by PES under the direction of Frank Phoenix. Mike Toney of EPA was on site and overallresponsibile for the testing.

1.4 Technical Systems Audit

The technical systems audit (TSA) was performed by R.K.M. Jayanty and Robert S. Wright ofResearch Triangle Institute (RTI) under EPA Contract 68-D4-0091, work assignment 99-03, from July


20, through July 26, 1998. The purpose of the audit was to conduct an independent technical assessmentof MRI and PES’s emissions testing through a technical review. The review included an in-depthevaluation of documents, on-site activities, equipment, procedures, record keeping, data validation, datamanagement, and reporting to ensure that established requirements are satisfied. The TSA wasconducted following principles described in a working draft version of EPA Guidance for TechnicalAssessments for Environmental Data Operations (EPA QA/G-7), which was being developed by EPA’sQuality Assurance Division at the time. Detailed findings and Technical Systems Audit Checklists asrevised by EPA for the tests performed by MRI and PES are presented in Chapters 2 and 3 respectively. Summaries of the findings and descriptions of the EPA revisions are presented below.

1.4.1 General Reviews

In general, RTI found that the PES and MRI team members performed the testing according tothe procedures outlined in the Site Specific Test Plan (SSTP) and Quality Assurance Project Plan(QAPP). Deviations from the QAPP or SSTP were discussed with the EPA Work Assignment Manager. RTI’s assessment proceeded after approval of the deviations. The equipment used was found by RTI tobe appropriate for the planned emissions testing and generally operated satisfactorily during the testing. RTI found that the PES and MRI team members, who were present at the site, to be well qualified andexperienced to perform the emissions testing and conducted themselves in a professional manner.

Although RTI’s assessment of the performance of the testing by MRI and PES is valid, a fewrevisions of their findings were made. While most of the revisions were made to clarify the finding oreliminate unnecessary comments, one revision was made that altered the determination of the potentialeffect on Data Quality. Some of the items in the TSA checklist for PES were not completed by RTI dueto lack of information available at the on-site operations. After the field test, RTI requested thisinformation from PES, but it was not received by RTI by the date of their report to EPA. The checklistwas completed based on the RTI auditors’ observations during sampling and their discussions with thePES task manager, sampling train operators, custodians, and PES quality assurance (QA) coordinator atthe site. Additional information on the calibration of the equipment in the laboratory prior to the on-sitetesting was requested, but had not been received by RTI. Examples of blank data sheets, custody forms,and labels have been supplied to the auditors and were included with the draft version of the checklist.

The following paragraphs describe the revisions that EPA made to the findings and technicalsystem audit checklists RTI submitted to EPA on September 29, 1998.

1.4.2 Technical Assessment of MRI’s Testing

RTI identified as a finding, which may have a potential effect on data, that besides all reportsgenerated by this project being reviewed formally by senior project personnel, quality control data shouldbe audited by the project’s quality assurance officer. While the letter RTI received from MRI did notaddress the auditing of quality control data, MRI’s standard procedures require a senior project person toreview raw test data and to randomly select data to be followed through the analysis and data processing. In addition, EPA performed a quality control audit of the data and reproduced the calculations from rawdata to final results. As a result, the clarifying statements by RTI were removed and this finding waschanged to one that is unlikely to have a effect on data quality.


In the findings which are unlikely to have an effect on data quality table, wording of the thirdfinding was changed to better characterize the level of detail used to describe the FTIR procedureswithout indicating the adequacy of the procedures. The complexity of the FTIR procedures precludes adescription as detailed and thorough as most other methods. In the Technical Systems Audit Checklist,sentences were removed from items 3 and 10 of the Quality System Documentation. In Item 3, thesentence indicating the QAPP and SSTP were not revised has been deleted. In Item 10, the sentence wascompleted by stating that the THC calibration data was recorded on a legal pad.

1.4.3 Technical Assessment of PES’s Testing

In the findings that may have a potential effect on data quality, items 1 and 2 were revised. Asentence was added to Item 1 explaining that this was a research study and that testing was not meant tohave validation. This was the first attempt to measure many of the pollutants at these sources. It wouldbe highly unusual and expensive to also attempt to validate the test methods at these sources withoutsome prior knowledge of the performance of some initial testing. The paragraph for Item 2 discussingthe spiking of surrogate compounds for the VOST cartridges has been completely revised to properlyreflect that the analytical laboratory spiked compounds before transport to the field-testing location. Thesentence identifying a better spiking methodology was removed.

In the findings that are unlikely to have an effect on data quality, Items 1 and 2 have beenrevised. In Item 1, the sentence identifying a better spiking methodology was removed. In Item 2 onesentence was revised and one was removed. This corrects an incorrect and contradictory statement inone sentence that a dedicated notebook was not maintained to record any problems or process changes.

In the Technical Systems Audit Checklist, a clarifying statement was added to the comment forItem 2 under General Quality Assurance Information indicating that key personnel provided sample trainoperators with information from the QAPP and SSTP as needed for their tasks.

This

is

a

Blank

Page

Chapter 2

Technical Systems AuditInstrumental Test Methods Performed byMidwest Research Institute atHot Mix Asphalt Plant C,Los Angeles, CA

This

is

a

Blank

Page

2.1Technical Systems Audit of Test Methods Performed by MRI

2.0 TECHNICAL SYSTEMS AUDIT - MIDWEST RESEARCH INSTITUTE

This chapter is the final technical systems audit (TSA) of the emissions testing performed byMidwest Research Institute (MRI) at Hot Mix Asphalt Plant C in the Los Angeles, California area. Asexplained in detail in Chapter 1, the final TSA includes revisions by the EPA Quality Assurance Officer. The draft TSA report from Research Triangle Institute (RTI) was delivered to the EPA QualityAssurance Officer on September 29, 1998. The TSA was performed by R.K.M. Jayanty and Robert S.Wright of RTI under EPA Contract 68-D4-0091, work assignment 99-03, from July 20, through July 26,1998. The purpose of the TSA was to conduct an independent technical assessment of MRI's emissionstesting through a technical review, which included an in-depth evaluation of documents, on-siteactivities, equipment, procedures, and record keeping to assure that quality assurance requirements weresatisfied. The TSA was conducted in accordance with principles described in the EPA QualityAssurance Division’s working draft version of EPA Guidance for Technical Assessments forEnvironmental Data Operations (EPA QA/G-7).

In general, MRI did a very good job during the emissions testing at the hot mix asphalt plant. The emissions testing was executed according to the quality assurance project plan (QAPP) and the site-specific test plan (SSTP) to a large degree. The equipment that was used is appropriate for the plannedemissions testing and it generally operated satisfactorily. The on-site project personnel were well-qualified to perform the emissions testing and conducted themselves in a professional manner.

ChecklistSection

Findings which may have a Potential Effect on Data Quality

G3, G31 The sample concentrator method for the FTIR instrument does not appear to havebeen throughly validated or throughly documented. In 1993, EntropyEnvironmentalists conducted a method validation at a coal-fired boiler, which waspublished as EPA Report No. EPA/454/R-95/004, July 1993 (NTIS Order No. PB95-193199INZ). This method validation found that EPA Method 301 validation criteriawere met for toluene and xylenes in concentrated samples. Pacific EnvironmentalServices found these compounds during its preliminary measurements at the hot-mixasphalt plant. If MRI discovers significant concentrations of other chemicalcompounds in the asphalt plant emissions, then the report of MRI's emissions testingshould note that the FTIR sample concentrator method has not been validated forthese additional compounds, which it does.

ChecklistSection

Findings which are Unlikely to have an Effect on Data Quality

A3 Some equipment and procedures that were used during the emissions testing in Julydiffer from those documented in the QAPP, which was submitted to EPA in March. In many cases, these changes are documented in the SSTP. Some changes weremade as recently as 3 weeks before the testing and are not documented in eitherdocument. Revisions and/or amendments to both documents should have occurred,according to procedures outlined within the QAPP.

ChecklistSection



A8 The total hydrocarbon (THC) analyzer's calibration data were recorded on a legalpad, rather that on a formal data sheet as was used for other measurements. Section5.1 of the QAPP states that information will be entered in standard data forms.

A14 A letter from MRI to RTI indicates that all reports that are generated by this projectwill be reviewed formally by senior project personnel. It is assumed these reports didreceive review by senior personnel.

A-Comments The QAPP does not describe the FTIR procedures in a lot of detail. It cites EPA TestMethod 320 and the FTIR Protocol, which are pretty general to describe the specificsof the data collection effort at the hot-mix asphalt plant.

F31 Test Method 320 specifies an accuracy of ±2% for the CTS. The CTS used for thisstudy had an accuracy of ±5%. The concentration of the CTS was not independentlyverified by MRI.

F-Comments QAPP Section 3.1.1 states that the time for 5 cell volumes to pass through the cell isconsidered the minimum interval separating independent samples. If the cell volumeis 8.5 L, then 5 cell volumes corresponds to 42.5 L. If the FTIR sample gas flow rateis 5 L/min and the FTIR sample interval is 2.5 minutes, only 12.5 L (or 1.5 cellvolumes) of sample gas passes through the cell between measurements. Therefore,an individual FTIR measurement cannot be considered to be completely independentof the measurements that immediately precede it. Scott Klamm confirms that it takes3 to 4 samples to flush the cell for CTS measurements.

F-Comments During the emissions testing, MRI analyzed a nine-component hydrocarboncalibration standard (Spectra Gases cylinder number CC91245) that had beenbrought to the hot mix asphalt plant by Emissions Monitoring, Inc. The analysis ofthis calibration standard does not constitute a performance evaluation of theextractive FTIR method because MRI was not informed of the analysis prior to theemissions testing nor during our initial on-site meeting on 20 July 1998. The resultsof the analysis were not available during the RTI assessor's conversation with MRI'sThomas Geyer on August 27, 1998. RTI suggests that the results of the MRI'sanalysis of this calibration standard be included in MRI's emissions testing report,which will allow EPA to compare the MRI analytical results with the attachedcertificate of analysis for the calibration standard.

G21 The SSTP indicates that a preheated vapor-phase surrogate spike will be loaded ontothe Tenax cartridge via the sampling probe. The surrogate spike was actually loadedin the monitoring trailer at room temperature.

G25 A duplicate train sample was collected on 25 July 1998, but the FTIR samplespectrum was not saved. A second attempt to collect a duplicate train sample wasscheduled for 27 July 1998 after the assessors departure. The results of the postsampling analysis of this sample were not available for review as of 27 August 1998.

ChecklistSection



H-Comments There appear to be some errors in the draft THC calibration data the MRI submittedfor review. The span drift data for stack dryer Runs 1 and 2 are identical, which is anunlikely occurrence. There appear to be errors in the calculated span drifts for DryerStack Run 1 and Load-out Run 1. These errors do not appear to be major problems.


Technical Systems Audit Checklist

Project: Emissions Testing at a Hot Mix Asphalt PlantOrganization: Midwest Research Institute (Kansas City, Missouri)Assessment Location: Asphalt Plant C, Los Angeles, CaliforniaAssessors: Robert S. Wright and R.K.M. Jayanty, Research Triangle Institute (Research

Triangle Park, North Carolina)Assessment Dates: July 20 through July 26, 1998

Brief Project Description: EPA is investigating hot mix asphalt plants to identify and quantify particulatematter and hazardous air pollutants (HAPs) emitted from asphalt cement load-out operations. EPA hasissued a work assignment to MRI to conduct an air emissions test program to collect data in support ofthe investigation. Asphalt Plant C in Los Angeles, California was chosen primarily because load-outemissions are controlled by a silo exhaust system and a load-out tunnel. The plant has a productioncapacity of more 650 tons per day. Approximately 2,000 tons per 4 hour period were produced, duringthe test. The primary objective of the project was to characterize air emissions of organic HAPs from thestorage silos, the load-out tunnel, and the hot mix dryer.

AUDIT QUESTIONSRESPONSE

COMMENTY N

NA

A. QUALITY SYSTEM DOCUMENTATION

1. Is there an approved qualityassurance project plan (QAPP) forthe project and has it beenreviewed by all appropriatepersonnel?

T A QAPP was submitted to EPA onMarch 27, 1998. Additionally, MRIsubmitted a site-specific test plan(SSTP) on June 22, 1998. Bothdocuments have been approved by EPA.

2. Is a copy of the QAPP maintainedat the field site? If not, brieflydescribe how and where QA andquality control (QC) requirementsand procedures are documented.

T Copies of the QAPP and SSTP areavailable in plain sight in the monitoringtrailer. No one has been observedconsulting these documents.


COMMENTY N

NA


3. Is the design and conduct of theproject as is specified in theQAPP?

T In general, the project is beingimplemented as was specified in theQAPP. However, some equipment andprocedures have changed since theQAPP was submitted to EPA. In manycases, these changes are documented inthe SSTP. Some changes were made asrecently as 3 weeks before the testingand are not documented. Revisionsand/or amendments to both documentsshould have occurred, according toprocedures outlined within the QAPP.

4. Are there deviations from theQAPP?

T There are some noticeable deviations,which do not appear to affect the qualityof the data being collected. Forexample, QAPP Section 3.1.2 states thatTenax will be spiked with an analyte orsurrogate compound during samplecollection if practical. The sampleconcentration description in the QAPPalso mentions a post-test, laboratorygaseous spiking procedure and otherprocedures. SSTP Section 5.1.3 statesthat a vapor phase spike will bepreheated and injected into the back ofthe sampling probe. However, thesurrogate spike gas was loaded onto theTenax trap in the trailer prior tosampling using the cool thermaldesorption system for the traps.


COMMENTY N

NA


5. How are any deviations from theQAPP noted?

T QAPP Section 6.3 indicates that theQAPP will be amended to correct minordiscrepancies that have no effect on theoverall conduct of the study. The QAPPpresents a form for documenting suchamendments. An entire chapter of theQAPP will be revised if major changesin the conduct of the study occur. However, no amendments or revisionshave been submitted to EPA. Anydeviations that arise before or during thetesting should be documented.

6. For each measured parameter, doesthe QAPP list the frequency ofcalibration, acceptance criteria forthe calibration, and the process forcalibration data reduction andreview?

T QAPP Table 2-1 and SSTP Table 5-2give the calibration frequencies andaccuracy and precision objectives forsome field test methods. SSTP Table 5-1 lists the calibration frequencies,acceptance limits, reference standards,and calibration techniques for othermethods.

7. Are written and approved standardoperating procedures (SOPs) usedin the project? If so, list them andnote whether they are available atthe field site. If not, brieflydescribe how and where the projectprocedures are documented.

T No SOPs were observed at the testingsite. The Fourier transform infrared(FTIR) sample concentration procedureis attached to the QAPP. EPA testmethods are cited and, in some cases,attached to the QAPP and SSTP (i.e.,Methods 25A and 320 and the FTIRProtocol).

8. Briefly describe how calibrationand other QC data are documented.

T Operational parameters and calibrationdata were recorded on paper data sheetsand Excel spreadsheets except for totalhydrocarbon (THC) analyzer'soperational parameters and calibrationdata, which were recorded on a legalpad.

9. Does the calibrationdocumentation show thatcalibrations are being performed atthe required frequency and in therequired manner??

T MRI calibration data indicates that theextractive FTIR instrument and the THCanalyzer were calibrated at the requiredfrequency and in the required manner.


COMMENTY N

NA


10. Are there standard paper orelectronic forms to recordcalibration data and operationalparameters?

T Standard paper data sheets and Excelspreadsheets were used for allmeasurements except the THCmeasurements, which was recorded on alegal pad.

11. Are the standard data forms dated? T The forms have a line for the entry ofthe date.

12. Is the person who recorded the dataidentified on the form?

T The forms have a line for the entry ofthe operator's name.

13. Are any paper records written inindelible ink?

T All reviewed data forms were filled outin ink.

14. Are the QC data reviewed byanother qualified person such asthe QA officer or the projectleader? Who is this individual?

T A letter from MRI to RTI indicates thatScott Klamm, Thomas Geyer, JohnHosenfeld, Bruce Diel, and JackBalsinger will review formally allreports generated by this project. However, there is no indication whetherthe raw QC data are reviewed directlyby any MRI personnel.

15. Is the project team adhering to theplanned schedule? If not, explainthe new schedule. Verify that allschedule changes have beenauthorized.

T The schedule is being followed to theextent allowed by unexpectedequipment breakdowns in the asphaltplant and FTIR instrumentationproblems. All schedule modificationswere authorized by the EPA WorkAssignment Manager.

Additional Questions or Comments: The QAPP does not describe the FTIR procedures in enoughdetail. It cites EPA Test Method 320 and the FTIR Protocol, which are too general to describe thespecifics of the data collection effort at the test site. For example, QAPP Section 2.2.1.1 states thatethylene in nitrogen will be used as the FTIR calibration transfer standard (CTS) and FTIR ProtocolSection 4.5 gives selection criteria for the CTS. However, nowhere is it stated why ethylene innitrogen was selected as the CTS for this project. A similar problem exists for selection of the Tenaxtrap surrogate spike gas.


COMMENTY N

NA


B. ORGANIZATION AND RESPONSIBILITIESIdentify the following personnel and determine whether they have the listed responsibilities:

1. MRI Work Assignment Leader:John Hosenfeld (Kansas City, MO)• responsible for overall

performance of the project, and• communications with EPA

T Mr. Hosenfeld was not present at theemissions testing.

2. MRI Quality Assurance Officer: Jack Balsinger (Kansas City, MO)

• prepare QAPPC review and monitor QA activities

T Mr. Balsinger was not present at theemissions testing.

3. MRI Project Task Leader:Scott Klamm (Kansas City, MO)C responsible for the on-site

emissions testing effortC supervision of all MRI on-site

and off-site staff • communication with other on-

site personnel.

T Mr. Klamm is also responsible for theFTIR instrument.

4. MRI FTIR Oversight:Thomas Geyer (RTP)C guidance for FTIR field data

collection• direct spectral analysis effort

T Mr. Geyer was not present at theemissions testing, but was available forconsultation by telephone. After testingwas completed, he analyzed the FTIRspectra to identify compounds andquantify concentrations.

5. MRI FTIR Operators:Scott Klamm and Andy Page• operation of FTIR instrument

and sample concentrator,• calibration of FTIR instrument

and sample concentrator, and• recording operational parameters

T Mr. Klamm has primary responsibilityfor operating the FTIR instrumentduring the emissions testing.

6. THC Analyzer Operators:Bob Gulick and Bobby Edwards• operation of THC analyzer,• calibration of THC analyzer, and• recording operational parameters

T Mr. Edwards is also a gas sampling trainoperator.


COMMENTY N

NA


7. MRI SF6 Tracer Gas Operator:Dan Neal• operation of SF6 tracer gas

release manifold

T Mr. Neal is also a gas sampling trainoperator.

8. Other MRI Testing Staff:Jim Surman and Pam Murowchick

T Mr. Surman is a gas sampling trainoperator. Ms. Murowchick collectsprocess monitor data in the asphalt plantcontrol room and alerts the SF6 tracergas operator of load-out tunnelactivities.

9. EPA Work Assignment Manager: Michael L. ToneyC oversight of emissions testing

and problem solving• communication with all on-site

personnel

T Mr. Toney was present throughout theentire testing program and coordinatedthe efforts of the emissions testingteams. He authorized all changes intesting schedule and procedures.

10. EPA QA Officer:Lara P. Autry• review and approve QAPP• review and approve QA activities

T Ms. Autry was not present at theemissions testing.

Additional Questions or Comments:


COMMENTY N

NA


C. TRAINING AND SAFETY

1. Does the FTIR instrument operatorhave special training or experiencefor the operation of the instrument?

T Mr. Klamm has 8 years experience withextractive FTIR and open-path FTIR inlaboratory and field applications.

2. Does the THC analyzer operatorhave special training or experiencefor the operation of the instrument?

T Mr. Gulick has 10 years experience withemissions testing and 5 additional yearsof experience with analyticalinstrumentation.

3. Does the project maintain currentsummaries of the training andqualifications of project personnel?

T A letter from MRI to RTI includes briefresumes for project personnel.

4. Does the project maintaindescriptions of assignmentresponsibilities?

T The QAPP contains descriptions ofsenior project personnel. A letter fromMRI to RTI states that theresponsibilities for other personnel wereassigned in meetings before and duringthe project.

5. Is there special safety equipmentrequired to ensure the health andsafety of project personnel?

T The asphalt plant required personnel towear hard hats while walking underconveyor belts.

6. Is each project team memberappropriately outfitted with safetygear?

T Each team member wore safety shoesand a hard hat when outside of thetrailer. Eye and ear protection wereworn on an as-needed basis.

7. Are project personnel adequatelytrained for their safety during theperformance of the project?

T Question not asked, but observation ofthe team members demonstrates thatthey were adequately trained aboutsafety.

8. Who is authorized to haltemissions testing in the event of ahealth or safety hazard?

T The EPA Work Assignment Managerretained authority to halt emissionstesting as necessary to protect healthand safety.



COMMENTY N

NA


D. DATA QUALITY INDICATOR GOALS AND PERFORMANCE TESTING

1. Is the anticipated use of the dataknown and documented in theQAPP?

T The study objectives are brieflydescribed in the QAPP.

2. What are the criticalmeasurements?

T The QAPP describes extractive Fouriertransform infrared (FTIR) spectroscopy,FTIR by sample concentration, and totalhydrocarbons (THC).

3. Have data quality indicator goalsfor each critical measurement beendocumented in the QAPP?

T QAPP Table 2-1 presents accuracy andprecision goals for direct and indirectFTIR, THC, moisture, temperature,velocity, oxygen, and carbon dioxide.

4. Do the above data quality indicatorgoals appear to be based ondocumented performance criteriafor the measured parameter or onactual QC data compiled for theparticular measured parameter?

T The accuracy and precision goals forextractive FTIR and THC are based onEPA Test Methods 320 and 25A. Theaccuracy and precision goals for FTIRby sample concentration are based onEPA Method 301 validation criteriafrom Entropy Environmentalists' 1993method validation at a coal-fired boiler.

5. Has the performance of each of thecritical measurements beenassessed and documented?

T Calibration data for direct and indirectFTIR and THC were obtained andrecorded on a daily basis.

6. Are there established proceduresfor corrective or response actionswhen measurement performancecriteria or the data quality indicatorgoals (e.g. out-of-controlcalibration data) are not met? Ifyes, briefly describe them.

T The corrective action procedures areinformal and are implemented on an as-needed basis. For example, problemswith the FTIR instrument prompted acall to a FTIR consultant, who was ableto provide suggestions for correctiveactions.

7. Are the corrective actionprocedures consistent with theQAPP?

T QAPP Section 4.6 describes generalcorrective action procedures that arefollowed when problems occur.


COMMENTY N

NA


8. Have any such corrective actionsbeen taken?

T A few minor corrective actions havebeen taken, but none were significantenough to halt work on the project or toinvolve the MRI work assignment leaderor the MRI quality assurance officer.


E. EXTRACTIVE SAMPLING SYSTEM

1. Describe locations of the samplingports.

T During the testing of the hot mix dryingexhaust, the sampling ports were locatedat the baghouse exhaust stackapproximately 10 feet below the outlet. The sample ports for the testing of theload-out tunnel were located in theexhaust ducting between Silos 1 and 2. The sampling ports for the testing of thesilos were located in the exhaust ductingof Silo 2 between the silo vent and thedamper.

2. Describe the sampling probes. T The sampling probes were short,straight lengths (i.e., 1 to 2 feet) ofstainless steel tubing that wereconnected to the particulate filters.

3. Are the sampling probes heated toprevent condensation?

T The sampling probes were locatedinside the sampling ports and remainedat that those temperatures. The dryerbaghouse exhaust temperature wasapproximately 260E C, the silo ducttemperature was approximately 240E C, and the load-out tunnel exhaust wasnear ambient air temperatures.

4. Does each of the sample probeshave a calibration valve assemblyfor sampling system bias tests?

T The sample probes have calibrationvalve assemblies for deliveringcalibration gas to the FTIR and THCinstruments via the sampling lines.


COMMENTY N

NA


5. Describe the sampling lines. T Two 100-foot lengths of TechnicalHeaters Model 22129-02-01-01sampling line containing three Teflon®tubes were used between the dryerbaghouse exhaust sampling port and thesampling pumps outside of the trailer. One tube conveyed sample gas, anothertube conveyed calibration gas, and thethird tube was not used. For the load-out tunnel and silo testing, one length ofsampling line was used between the siloand load-out sampling ports and anotherwas used between the load-out tunnelsampling ports and the sampling pumps. Two Teflon® tubes conveyed samplegas and the third tube conveyedcalibration gas. A Furon UnithermModel 220-666 sampling line containingthree Teflon® tubes was used betweenthe sampling pumps and the sampledistribution manifold inside the trailer.

6. Are the sampling lines heated toprevent condensation?

T The sampling line is heated to 300E Cusing a control and display box in thetrailer.

7. Describe any sampleconditioning/moistureremoval/dilution air system.

T There was no sample conditioning ordilution system.

8. Describe how the instrumentoperators switch from samplingfrom one location to sampling toanother location.

T FTIR and THC instrument operatorsswitch between sampling lines usingvalves in the sample distributionmanifold.

9. Is any sample conditioning systemmaintained according to schedule?

T Not applicable.

10. Describe the particulate filter. T A Balston particulate filter is containedin a stainless steel housing.


COMMENTY N

NA


11. Is the filter changed according toschedule?

T Question not asked. There did notappear to be enough particulate matterin the sample streams to requirechanging the filter.

12. Describe the sample pump. T Two KNF Model UN035ST.11Idiaphragm pumps with stainless steelheads.

13. Describe the sample flow controlapparatus.

T The sample flow rate was controlled bya shutoff valve in the sampledistribution manifold and it was 12L/min. The FTIR instrument required asample flow rate of 5 L/min and theTHC analyzer required a flow rate of2.5 L/min. All three flow rates wereindicated by a flow meter in themanifold. These flow rates were notcritical measurements and were notrecorded on data sheets. A back-pressure regulator maintained themanifold pressure at 4 to 5 psig. Excesssample not used by the two instrumentswas discarded through the sampledistribution manifold vent.

14. Describe how the total samplevolume is measured.

T Not applicable.

15. Describe how the sample volumemeter is calibrated.

T Not applicable.

16. When was the last time that thesample volume meter wascalibrated?

T Not applicable.

17. Is the volumetric calibrationtraceable to NIST standards?

T Not applicable.

18. How is the gas meter temperaturemeasured?

T Not applicable.

19. How is the gas meter thermometercalibrated?

T Not applicable.


COMMENTY N

NA


20. When was the last time that the gasmeter thermometer was calibrated?

T Not applicable.

21. Is the thermometer calibrationtraceable to NIST standards?

T Not applicable.

22. How is the stack gas streamtemperature measured?

T A Type-K thermocouple was mountedon the Pitot tube and was measured withan Omega Model HH81 digitalthermometer.

23. How is the temperature sensorcalibrated?

T The thermocouple was calibrated in aboiling water bath, an ice water bath, ora newly-acquired dry well and thecalibration temperature was measuredwith an ASTM mercury-in-glassthermometer.

24. When was the last time that thetemperature sensor was calibrated?

T 24 March 1998


T ASTM thermometers are traceable toNIST standards.

26. How is barometric pressuremeasured?

T Taylor aneroid barometer (Asheville,NC) with a 600-foot altitude correction. The asphalt plant is 540 feet abovemean sea level.

27. How is the barometer calibrated? T The aneroid barometer is compared to aSargent-Welch Model S-4519 mercurycolumn barometer.

28. When was the last time that thebarometer was calibrated?

T 14 July 1998

29. Is the barometric pressurecalibration traceable to NISTstandards?

T The mercury column barometer isgenerally considered to be a primarystandard.


COMMENTY N

NA


30. Describe the sample manifold. T Sample gas enters the sampledistribution manifold where it isdistributed by switching valves, shutoffvalves, and flow meters to the FTIR andTHC instruments. A back-pressureregulator maintains the pressure in themanifold. Excess flow is vented outsidethe trailer.

31. Describe the calibration gasmanifold and associated calibrationgas lines to the sample probe andsample manifold.

T Calibration gas enters the sampledistribution manifold where it isdistributed by switching valves, needlevalves and flow meters either directly tothe FTIR and THC instruments orindirectly to them via the sampling linesand the calibration valves on thesampling probes.


F. EXTRACTIVE FOURIER TRANSFORM INFRARED (FTIR) INSTRUMENT

1. Describe the extractive FTIRinstrument. List the brand, modelnumber, and serial number.

T KVB Analect (Irvine, CA) Model RFX-40 FTIR instrument with a glow-barlight source and a liquid-nitrogen-cooled mercury-cadmium-telluride(MCT) detector. The FTIR instrumenthas a 1 cm-1 resolution, 400 to 4400 cm-

1 spectral range.

2. Does the instrument operateaccording to an EPA method?

T EPA Test Method 320


COMMENTY N

NA


3. Describe the gas containment cell. T Infrared Analysis Model D-22H variablepathlength White cell inside an InfraredAnalysis G-5-22-V-BA-AU heated gascell oven that was maintained at atemperature of 250E C. The light pathwas aligned with a laser. Thepathlength was approximately 9 to 10meters and was determined on a dailybasis by measurement of the calibrationtransfer standard (CTS).

4. Is the pressure in the gascontainment cell monitored?

T The cell pressure was monitored with anEdwards Model W60041111 Barocelpressure sensor with a 0 to 1000 mm Hgrange. This sensor is calibratedannually at the factory and its readingswere verified using the barometer.

5. Describe the sample lines betweenthe sample distribution manifoldand the gas cell.

T A 20-foot length of Technical Heaterssampling line connected the FTIRinstrument to the sample distributionmanifold. The sampling line wasmaintained at a temperature of 300E F.

6. How are FTIR data recorded (e.g.,data acquisition system)? Brieflydescribe the system, giving itsbrand, model, and serial number.

T Interferograms and sample spectra wererecorded on the hard drive of thepersonal computer running the FTIRsoftware and on an external Iomega Jazdrive.

7. Does the data recording systemhave a provision for documentingoperating parameters (e.g.,resolution, pathlength, scannumber, sampling time, etc.) forindividual spectra? If not, arethese parameters documented insome other manner?

T The data recording system recordedparameters (e.g., resolution) that weredirectly related to the operation of theFTIR instrument. Parameters (e.g.,pathlength) that were not related to theoperation of the FTIR instrument wererecorded on a paper data sheet.

8. Is there a back-up for the datarecording system?

T The external Iomega Jaz drive was theback-up data recording system.


COMMENTY N

NA


9. Can raw FTIR spectra be recoveredfrom the backup?

T Interferograms and sample spectra couldbe recovered from the back-up datarecording system.

10. Describe the routine sampling andanalysis procedure for theinstrument.

T Sample gas flowed through the gascontainment cell on a continuous basis. The sample flow rate was 5 liters perminute. Over a 2- to 2.5-minutesampling interval, 100 interferometerscans were coadded. The FTIRinstrument converted the coaddedInterferograms into sample spectra.

11. What is the leak volume for the gascontainment cell?

T Daily leak checks indicate that the cellleak rate meets the acceptance criterionof less than 4 %/minute.

12. What is the noise level in eachanalytical region?

T Typical noise levels range between 10-3

and 10-4. On the morning of 22 July 98,the noise level was 10-4 in the 1000 to1300 cm-1 spectral region in whichvolatile organic compounds (VOC)absorb light. In general, noise levels aredependent on sample moisture levels,which will vary from source to source.

13. What is the sample absorptionpathlength for each analyticalregion?

The nominal pathlength is 10.5 meters. The actual pathlength was determinedfrom measurements of the CTS. SeeItem G35 below.

14. What is the fractional analysisuncertainty for each analyticalregion?

The calculated uncertainty in themeasured concentration of the CTS because of errors in the mathematicalcomparison of reference and samplespectra will be determined during thepost sampling analysis phase of theproject .

15. What is the calibration frequency? T The CTS was measured at the beginningand end of each test. The sulfurhexafluoride (SF6) calibration gas andthe SF6/toluene sampling bias check gaswere measured daily.


COMMENTY N

NA


16. Describe the routine calibrationprocedure for the instrument.

T The gas containment cell was evacuatedto a pressure of 10 mm Hg and then wasfilled to atmospheric pressure with theCTS. The mean CTS concentrationacross the entire project was comparedwith the certified value to determine thepathlength of the cell. The SF6 standardwas used for the capture efficiency tests. The SF6/toluene sampling bias checkwas used to determine sampling linelosses by comparing the ratio of direct FTIR measurements of the twocompounds in the gas mixture with thecorresponding ratio for measurements ofthe same gas mixture after it passedthrough the sample probe calibrationvalve and the sampling line.

17. Does the calibrationdocumentation show that thecalibration procedures are beingfollowed?

A letter from MRI to RTI presents CTScalibration data which were collectedseveral times each day. No SF6 orSF6/toluene calibration data werepresented in the letter because they willbe determined during the post samplinganalysis phase of the project.

18. What is the sampling system biasaccording to the calibrationdocumentation?

The SF6/toluene calibration mixture isdelivered to the calibration valve on thesample probe and directly to the FTIRinstrument. The bias will be determinedduring the post sampling analysis phaseof the project by comparison of the twosets of SF6/toluene data.

19. How frequently are backgroundspectra collected?

T Background spectra using pure nitrogenwere collected at the beginning and endof the day. The first spectrum was usedroutinely for the entire day unless thesecond spectrum seemed to be morerepresentative.


COMMENTY N

NA


20. Are emissions measurementscorrected for backgroundinterference?

T Sample spectra were corrected forbackground interferences associatedwith the light source, the gas cell, andthe detector.

21. How are compounds in the sampleidentified?

T Thomas Geyer will analyze samplespectra and will identify detectedcompounds after the completion of thetesting using reference spectra or otherspectroscopic analysis techniques.

22. What is the source of referencespectra and absorption coefficientsfor identified compounds?

T The EPA reference spectra database willbe the primary source of referencespectra. Other spectral databases willbe used if necessary.

23. What will be done for compoundswithout reference spectra?

T If MRI discovers significant,unidentified spectral features in thesample spectra, the EPA WorkAssignment Manager will be consultedregarding whether these features are tobe identified by laboratorymeasurements.

24. How will unidentified peaks bereported?

T The primary goal of the testing was notto identify all compounds in the samplespectra, but to determine theconcentrations in the emissions of aspecific list of hazardous air pollutants(HAPs) and other compounds.

25. How are concentrationscalculated?

T QAPP Section 4.4 states that a K-matrixanalytical procedure used uses samplespectra collected during the emissionstesting and reference spectra foridentified compounds from the EPAlibrary to calculate a least-squares fit ofthe spectral features and to determine aconcentration for each identifiedcompound.


COMMENTY N

NA


26. What are the minimum detectableconcentration for the compoundsof interest?

T Scott Klamm estimated that theminimum detectable concentrationranges between 0.5 and 5 ppmdepending on the compound. ThomasGeyer noted that the sensitivity limit isat low ppb-levels and the minimumquantitation limit (MQL) is in low ppm-levels for samples containingconsiderable amounts of water andcarbon dioxide. For toluene and thexylenes, the MQL is estimated to begreater than the 100 ppb that wasmeasured during preliminary testing.

27. Does the FTIR instrument haveany spectral interferants for thecompounds of interest?

T Water vapor and carbon dioxide arecommon analytical interferants insample spectra.

28. How are the FTIR data correctedfor analytical interferants?

T An integrated bag sample of theemissions was collected for each testrun. The bag's moisture and carbondioxide levels were determined with anOrsat analyzer. The measured levelswere used to select appropriate watervapor and carbon dioxide referencespectra for use in sample spectraanalysis by Thomas Geyer aftercompletion of the field portion of theproject.


COMMENTY N

NA


29. List the calibration transferstandards (CTSs) for the FTIRinstrument

T The CTS was Scott Specialty Gasescylinder number ALM005893containing 99.9 parts per million (ppm)ethylene in nitrogen. This gas mixturewas analyzed on 31 March 1998.

The SF6 tracer calibration gas was ScottSpecialty Gases cylinder numberALM033887 containing 0.205 ppm SF6in nitrogen. It was analyzed 7 April1998.

The sampling bias check gas was ScottSpecialty Gases cylinder numberAAL17264 containing 3.83 ppm SF6and 105 ppm toluene in nitrogen. It wasanalyzed on 7 April 1998.

30. Do the CTSs have appropriate gasmixtures and concentrations for thesample gas mixtures andconcentrations?

FTIR Protocol Section 4.5.1 specifiesthat each analytical region lie within25% of the CTS position. The CTS forthis testing is a compressed gascalibration standard containing 100 ppm(±2 %) ethylene in nitrogen. This gasmixture was selected as the CTSbecause it has an light absorption line at949 cm-1, which meets the specificationfor SF6 (942 cm-1), toluene (727 cm-1),and xylenes (740 to 797 cm-1).

31. What is the analytical uncertaintyof the CTSs?

T A letter from MRI to RTI presentscertificates of analysis for CTS, SF6,and SF6/toluene calibration standards. All three standards have an analyticaluncertainty of ±5%. Section 7.2 of TestMethod 320 directs the analyst to"obtain a NIST-traceable, gravimetricstandard of the CTS (±2 percent)".


COMMENTY N

NA


32. Are the CTSs traceable to NISTSRMs or otherwise traceable toNIST? EPA Test Method 320specifies that the CTS be a NIST-traceable gravimetric standard (±2%).

T A letter from MRI to RTI states that thecalibration standards are traceable to thespecialty gas producer's primaryreference standards, which are preparedusing gravimetric procedures. Thesecalibration standards are not traceable toNIST as NIST defines traceability forcompressed gas standards. NIST doesnot have SRMs for this gas mixture.

33. Have the CTSs been independentlyverified by MRI?

T The calibration standards were notindependently verified by MRI.

34. Are the CTS regulators anddelivery system properlymaintained?

T The pressure regulators and theassociated gas handling equipmentappear to be well-maintained.

35. What is the variation of successiveCTS absorbency measurementsrelative to their mean value?

T A letter from MRI to RTI presents theCTS calibration data, which yielded amean pathlength of 10.475 meters and amaximum deviation from the mean of0.337 meters or 3.2% of the mean. These data meet the QAPP's precisionspecification of agreement to within±%5 of the mean.

36. Is there a schedule of preventivemaintenance for the FTIRinstrument?

T The instrument is three years old andScott Klamm indicated that it had notbeen serviced by a factoryrepresentative during that period. Allmaintenance is done on an as-neededbasis.

37. Are calibration and maintenancelogs kept for the FTIR instrument?

T Calibration data are recorded on datasheets, rather than in logbooks. Thereare no maintenance logs.


COMMENTY N

NA


38. Review any maintenance andoperational records for the FTIRinstrument. Based on theserecords, does the instrument appearto be in good operating condition?

T The FTIR instrument was manufacturedby KVB Analect, which is located inIrvine, California. A factory servicerepresentative optimized theinterferometer on 20 July 1998 in thepresence of the assessors. The FTIRinstrument appeared to be in goodoperating condition after it wasoptimized.

39. Are the manufacturer’s operatingmanuals readily available to theFTIR instrument operators?

T Neither MRI's FTIR operator nor theKVB Analect factory servicerepresentative had a copy of theoperating manual.

Additional Questions or Comments: QAPP Section 3.1.1 states that the time for 5 cell volumes topass through the cell is considered the minimum interval separating independent samples. If the cellvolume is 8.5 L, then 5 cell volumes corresponds to 42.5 L. If the FTIR sample gas flow rate is 5L/min and the FTIR sample interval is 2.5 minutes, only 12.5 L (or 1.5 cell volumes) of sample gaspasses through the cell between measurements. Therefore, a individual FTIR measurement cannot beconsidered to be completely independent of the measurements that immediately precede it. ScottKlamm confirms that it takes 3 to 4 samples to flush the cell for CTS measurements.

During the emissions testing, MRI analyzed a nine-component hydrocarbon calibration standard(Spectra Gases cylinder number CC91245) that had been brought to the hot mix asphalt plant byEmissions Monitoring, Inc. The analysis of this calibration standard does not constitute aperformance evaluation of the extractive FTIR method because MRI was not informed of the analysisprior to the emissions testing nor during our initial on-site meeting on 20 July 1998. The results ofthe analysis were not available during the RTI assessor's conversation with MRI's Thomas Geyer onAugust 27, 1998. RTI suggests that the results of the MRI's analysis of this calibration standard beincluded in MRI's emissions testing report, which will allow EPA to compare the MRI analyticalresults with the attached certificate of analysis for the calibration standard.


COMMENTY N

NA


G. SAMPLE CONCENTRATOR FOR FTIR INSTRUMENT

1. Describe the sample concentrator. T Samples were concentrated for FTIRanalysis by pulling source gas throughtwo traps containing 20 grams of Tenaxadsorbent. VOCs that were collected onthe Tenax were desorbed by heating thetrap to 220E C. Nitrogen flowingthrough the heated trap transferred theVOCs to an evacuated gas containmentcell, which has a volume of 8.5 L. If thesource gas flowed through the Tenaxtrap for 4 hours at a flow rate of 1.5L/min, source gas samples wereconcentrated by a factor of about 42.

2. Does the concentrator operateaccording to an EPA method?

T The method is still experimental and hasnot been standardized as an EPA testmethod.

3. If it is not an EPA method, whatdocumentation exists concerninglaboratory and field validations ofthe concentrator?

See comment below Item G32.


COMMENTY N

NA


4. Have the validations involved thecompounds of interest to thisstudy?

T SSTP Section 4.1.2 states that preliminary measurements by PacificEnvironmental Services (PES) at theasphalt plant indicated the presence oftoluene and xylenes at concentrationsbelow 100 ppb.

The 1993 FTIR method validationreport by Entropy Environmentalistsincluded the following correctionfactors and relative standard deviations(RSDs) in concentrated samples:

Correction Compound Factor RSD(%) toluene 0.83 10.31 m-xylene 0.78 4.85 p-xylene 1.16 14.25

For the method to be acceptable, thecorrection factor must be between 0.70and 1.30 and the RSD must be < 50%.

MRI does not plan to correct theindirect FTIR data for the surrogatespike gas collection efficiency.

5. Describe the sample lines betweenthe sample distribution manifoldand the concentrator.

T The sample concentrator was notconnected to the sample distributionmanifold. It was mounted on a separateVOST sampling train at the samplingport.

6. Describe how sample is drawnthrough the Tenax adsorbent trap.

T Source gas is pulled through thesampling probe, a 5-foot run of heatedsampling line, an intake manifold, acondenser, one Tenax trap, a waterdrop-out trap, a second condenser, asecond Tenax trap, and a Nutech Model280/01BVOST pump box.


COMMENTY N

NA


7. What are the sample flow rate,sample duration, and samplevolume?

T The sample flow rate was 1.5 L/min, thesample duration was 4 hours, and thesample volume was 360 L.

8. Describe how the flow rate ismeasured and controlled.

T A Matheson rotameter measured thesample flow rate and a needle valvecontrolled the flow rate.

9. Describe how the total samplevolume is measured.

T A Single Model 802 dry gas meter in theVOST pump box measured the samplevolume.

10. Describe how the sample volumemeter is calibrated.

T The dry gas meter and the rotameterwere calibrated by a wet test meter inthree separate runs of 14 to 21 L each

11. When was the last time that thesample volume meter wascalibrated?

T 13 July 1998

12. Is the volumetric calibrationtraceable to NIST standards?

The wet test meter was calibrated on anannual basis with a calibrated watervolume, which MRI considers to be aprimary standard. A wet test metercalibration data sheet dated 2 May 1997shows that it was calibrated according toASTM Method D1071-83.

13. How is the first condenser'stemperature measured?

T The temperature was measured with aType-K thermocouple and a OmegaModel HH81 digital thermometer.

14. How is the first condenser'sthermometer calibrated?

T The thermocouple was calibrated with aboiling water bath, an ice water bath, ora recently-acquired dry well asappropriate. An ASTM mercury-in-glass thermometer was the referencestandard.

15. When was the last time that thefirst condenser's thermometer wascalibrated?

T 24 March 1998


T ASTM thermometers are traceable toNIST.


COMMENTY N

NA


17. How is Tenax cleaned prior tosampling?

T Precleaned Tenax was purchased for theemissions testing. Before a Tenax trapwas used, it was further cleaned bybeing heated in the thermal desorber to180E C for 1 hour while nitrogenremoved any residual VOCs. TheTenax trap was then desorbed into theFTIR instrument and a blank spectrumfor the cleaned trap was recorded.

18. Has the Tenax used in this projectpassed the 5-ppb THC pass/failcriterion for cleanliness?

T Cleanliness for cleaned Tenax traps wasverified by field FTIR checks, ratherthan by laboratory FID checks, whichare associated with the 5 ppb cleanlinesscriterion.

19. What is the surrogate spike gas? T Scott Specialty Gases cylinder numberALM031809, 10.6 ppm toluene-d8 innitrogen, analyzed 2 April 1998. Theattached certificate of analysis lists theanalytical accuracy of ±5%. Section 9.1of Test Method 320 specifies the use of"a certified standard (accurate to ±2percent) of the target analyte, if one canbe obtained".


COMMENTY N

NA


20. Does the surrogate spike gas havean appropriate gas mixture andconcentration for the sample gasmixture and concentration?

T SSTP Section 5.2 states that toluene waschosen as the surrogate spiking gasbased on results reported by PES frompreliminary sampling conducted duringtheir pretest site survey. Thesemeasurements indicated the presence oftoluene and xylenes at concentrationsbelow 100 ppb. Item G1 abovedetermines the Tenax concentrationfactor to be 42. If the sampleconcentration method had been usedduring PES' preliminary measurements,the toluene concentration in the gas cellwould have been less than 4.2 ppm. This calculated concentration differsfrom the surrogate spiking gasconcentration by a factor of 2.5. Appendix D in the QAPP states thatsurrogate spiking gas concentrationsshould approximate the levels expectedin the trap after sample collection.

21. How is the surrogate spike gasadded to the sample?

T Approximately 1 gas cell volume of thesurrogate spike gas was loaded onto theTenax trap before sampling using thethermal desorber at ambienttemperature. Note that this spikingprocedure deviated from QAPP Section5.1.3, which states that the vapor-phasespike will be heated and injected intothe back of the sampling probe, similarto the Method 320 analyte spikeprocedure. This deviation wasnecessary because a VOST train wasused to pull source gas directly from thesource through about 5 feet of heatedsample line into the Tenax trap.


COMMENTY N

NA


22. Describe any ambient air samplesthat are collected.

T An ambient air sample was collected onthe morning of Wednesday, 22 July1998 using VOST Control Box #3. Thesample was collected at the load-outtunnel entrance where diesel trucks waitto be loaded.

23. Describe any train blanks that arecollected.

T A train blank was collected in the trailerat the beginning of the emissions testingby passing 90 L (1.5 L/min for 1 hour)of nitrogen through the Tenax train

24. Describe any preliminary samplingthat occurs.

T No preliminary sampling was conducteddue to time delays at the beginning ofthe emissions testing due to asphaltplant equipment breakdowns.

25. Describe any duplicate trainsamples that are collected.

T A duplicate train sample was collectedon 25 July 1998, but the FTIR samplespectrum was not saved. A secondattempt to collect a duplicate trainsample was scheduled for 27 July 1998after the assessors departure. Theresults of the post sampling analysis ofthis sample were not available forreview as of 27 August 1998.

26. Describe any breakthrough trapsthat are used in the sample train.

T A second Tenax trap was mounteddownstream of the first trap to collectany VOCs that might break through.

27. Describe how Tenax samples arestored prior to analysis.

T Tenax traps were capped after samplingand then were stored on ice until theywere analyzed.

28. How much time elapses betweensampling and analysis of Tenaxsamples?

T In general, the Tenax traps are analyzedwithin an hour or two of the end of atest run. However, the traps of 24 July1998 were not analyzed until 25 July1998 due to the need to produce cleanedtraps for the next day.


COMMENTY N

NA


29. How are Tenax samples identified? T All Tenax traps had an engravednumber on their bodies. Additionally,green sticky tape with the intendedusage was attached to the traps aftercleaning.

30. How is the sample desorbed fromthe Tenax and transferred to thegas containment cell?

T The FTIR gas containment cell wasevacuated to a pressure of 10 mm Hg. The Tenax trap was heated to 220E C inthe thermal desorber. A heated transferline at 250E C was opened between thethermal desorber. Nitrogen carried theVOCs into the 8.5-L gas cell at a flowrate of 1 L/min until the cell returned toatmospheric pressure.

31. What current or previous data areavailable concerning thedesorption efficiencies for thecompounds of interest and thesurrogate gas?

T The 1993 FTIR method validationreport by Entropy Environmentalistsincluded the following correctionfactors:

toluene 0.83 m-xylene 0.78 p-xylene 1.16

For the method to be acceptable, thecorrection factor must be between 0.70and 1.30 as per EPA Method 301. Noinformation is available for thedesorption efficiencies for othercompounds that might be found in theasphalt plant.

32. What is the detection limit for thecompounds of interest using thesample concentrator?

T The attached letter from MRI impliesthat sampling parameters were chosen toobtain minimum detectableconcentrations of 100 ppb for tolueneand xylene, which were identified byPES' preliminary measurements.


COMMENTY N

NA


Additional Questions or Comments: The sample for the morning of 22 July 1998 had an ambientair leak for an unknown period of time (10 minutes?) due to the first condenser separating from theinlet manifold.

There is little documentation for the FTIR sample concentration method. The QAPP contains a fairlygeneral discussion of the method that only briefly mentions the validations of the method. Thevalidations were reported to EPA in project reports by Entropy Environmentalists and MRI. In 1993,Entropy performed a FTIR method validation at a coal-fired boiler. Multi component gas mixtureswere spiked into boiler emissions, which were sampled by four parallel sample concentrators similarto the one used in this emissions testing.

This method validation found that EPA Method 301 validation criteria were met for toluene andxylenes in concentrated samples. Pacific Environmental Services found these compounds during itspreliminary measurements at the hot-mix asphalt plant. If MRI discovers significant concentrationsof other chemical compounds in the asphalt plant emissions, then the report of MRI's emissionstesting should note that the FTIR sample concentrator method has not been validated for theseadditional compounds.

The 1993 tests used a KVB Analect Model RFX-40 FTIR instrument with a MCT detector and anInfrared Analysis Model 5-22H gas containment cell with a pathlength of 22 m. The sample volumewas 280 L. The target gas mixture concentrations in the FTIR gas cell were 20 ppm. The CTS was100 ppm ethylene in nitrogen.

H. TOTAL HYDROCARBONS (THC) ANALYZER

1. Describe the THC analyzer. Listthe brand, model number, andserial number.

T Two J.U.M. Engineering Model VE7total hydrocarbon analyzers were usedfor the emissions testing. A thirdanalyzer was rented when one of theoriginal analyzers failed.

2. Does the THC analyzer operateaccording to an EPA method?

T EPA Test Method 25A

3. How are THC data recorded (e.g.,data acquisition system)? Brieflydescribe the system, giving itsbrand, model, and serial number.

T THC data were recorded with aWinbook XP laptop computer with adata acquisition system docking stationand Labtech Notebook software. Datawere collected at 1-second intervals and1-minute mean concentrations werestored on the computers hard drive.


COMMENTY N

NA


4. Does the THC data recordingsystem have a provision fordocumenting changes in operatingparameters? If not, are changes inoperating parameters documentedin some other manner?

T The data recording system collected thedate and time, but not other operationalparameters. Other parameters,including calibration data, wererecorded on a legal pad.

5. Is there a hard copy back-up for theTHC data recording system?

T A paper copy of data from the computeris printed by a Panasonic KX-P1180iprinter.

6. Can THC data be recovered fromthe hard copy backup?

T Data would have to be reduced by handfrom the paper copy.

7. Describe the sample lines betweenthe sample distribution manifoldand the THC analyzer.

T Approximately 6 feet of heatedsampling line connect the sampledistribution manifold to the THCanalyzer. The sampling line wasmaintained at 300E C.

8. Describe how sample is drawnthrough the THC analyzer.

T A pump pulled sample into an ovenheated to 300E C and through a samplefilter at a flow rate of approximately 2.5L/min. The pump pushed samplethrough a sample capillary into theflame ionization detector (FID). Theexcess sample was vented through abypass capillary. A back-pressureregulator and a pressure gauge controlsthe sample pressure to approximately200 millibar (or 3 psig) inside theinstrument.

9. What is the sample flow rate andthe bypass flow rate?

T The total sample flow rate wasapproximately 2.5 L/min and the flowrate into the FID was approximately 20mL/min.

10. Describe how the flow rates aremeasured and controlled.

T The flow rates are controlled by thecapillaries and the back-pressureregulator. The flow rates are notmeasured.

11. Describe how the flow rate metersis calibrated.

T Not applicable.


COMMENTY N

NA


12. When was the last time that theflow rate meters were calibrated?

T Not applicable.

13. List the flame ionization detectorhydrogen (vendor, grade, etc.).

T Air Products cylinder numberSGKK064 (40%hydrogen, 60% helium,analyzed on 11 June 1998).

14. List the flame ionization detectorburner air (vendor, grade, etc.).

T Ambient air is drawn through a charcoalscrubber by an internal pump andhydrocarbons in the air are removed. Bob Gulick indicated that methanepasses through the scrubber.

15. What is the THC analyzercalibration frequency?

T As per Method 25A, the THC analyzeris calibrated at the beginning of the testrun and a drift check is done at the endof the run. Bob Gulick will do morefrequent drift checks if he is concernedabout possible drift during a test run.

16. Describe the routine THC analyzercalibration procedure.

T Calibration gas mixtures of varyingconcentrations (e.g., 0, 25, 50, and 90ppm) are prepared in the EnvironicsModel 2020 gas dilution system from apropane calibration standard andnitrogen. The calibration gas mixturesflow through the sample line to thecalibration valve in the sample probe.

17. Does the THC analyzer calibrationdocumentation show that thecalibration procedures are beingfollowed?

T The attached letter from MRI presentsdraft calibration data for the THCanalyzer, which indicate that thecalibration procedures were followed.

18. List the THC calibration gases(including zero gas).

T Air Products and Chemicals cylindernumber SG9168085, 3690 ± 23 ppmpropane in nitrogen (EPA ProtocolGas), analyzed by gas chromatography-FID on 5 August 1996.


COMMENTY N

NA


19. Do the THC calibration gases haveappropriate gas mixtures andconcentrations for the sample gasmixtures and concentrations?

T The THC calibration gas mixture isappropriate. The concentration isappropriate because the gas mixture isdiluted by an Environics Model 2020gas dilution system, whose flow meterswere calibrated at the factory using aSierra Cal Bench on 23 April 1998.

20. What is the analytical uncertaintyof the THC calibration gases?

T The analytical uncertainty of the THCcalibration gas is 23 ppm or 0.6 percentof the certified concentration. Thespecified uncertainty of the gas dilutionsystem's flow rates is 0.5 percent.

21. Are the THC calibration gasestraceable to NIST SRMs orotherwise traceable to NIST?

T The THC calibration gas is traceable viaa 4723 ppm propane gas manufacturer'sinternal standard (GMIS), which isdirectly traceable to a NIST SRM.

22. Have the THC calibration gasesbeen independently verified byMRI?

T An independent verification is notnecessary because the THC calibrationgas is an EPA Protocol Gas.

23. Are the THC calibration gasregulators and delivery systemproperly maintained?

T No problems were observed.

24. What is the THC analyzercalibration error according to thecalibration documentation?

T The attached letter from MRI presentsdraft calibration data for the THCanalyzer, which indicate that thecalibration error ranged between 0.0%and 4.8%, which is less than the ±5%criterion specified in Test Method 25A.

25. What is the THC analyzer linearerror according to the calibrationdocumentation?

T Section 2.2.1.2 of the QAPP specifies a±2% accuracy criterion for linear. However, this criterion does not appearin Test Method 25A and appears to bean error in the QAPP.


COMMENTY N

NA


26. What are the THC analyzer zeroand calibration drifts according tothe calibration documentation?

T The attached letter from MRI presentsdraft calibration data for the THCanalyzer, which indicate that zero driftranged between 0.0% and 1.0%, whichis less than the ±3% criterion specifiedin Test Method 25A. Span drift rangedbetween 0.0% and 1.4%, which is lessthan the ±3% criterion.

27. What is the sampling system biasaccording to the THC analyzercalibration documentation?

T Section 2.2.1.2 of the QAPP specifies a±5% accuracy criterion for systems bias. However this criterion does not appearin Test Method 25A and appears to bean error in the QAPP.

28. Is there a schedule of preventivemaintenance for the THCanalyzer?

T Maintenance is performed on an as-needed basis in MRI's laboratories andin the field.

29. Are calibration and maintenancelogs kept for the THC analyzer?

T No logs are available.

30. Review maintenance andoperational records for the THCanalyzer. Based on your findings,does it appear to be in goodoperating condition?

T Visual Inspection of the two THCanalyzers indicates that they were ingood operating condition at thebeginning of the project. See additionalcomment below.

31. Are the manufacturer’s operatingmanuals readily available to theTHC analyzer operator?

T The operator has a manual for the THCanalyzer.


COMMENTY N

NA


Additional Questions or Comments: Something associated with emissions from the silo seemed tobe damaging THC analyzers. Beginning on 23 July 1998, one THC analyzer sampled the load-outtunnel and another THC analyzer sampler the silo. On 24 July 1998, the silo THC analyzer's FIDflamed out and could not be relighted. The remaining THC analyzer sampled both locations untilearly afternoon on 25 July 1998 when it also flamed out. A rental THC analyzer arrived at noon on25 July 1998 and was used for the remainder of the day. The second THC analyzer's FID was baked at high temperature and appeared to be operational on 26 July 1998, although it was not used for thatday's sampling. One hypothesis is that heavy hydrocarbons from the silo or their oxidation productsmay be clogging the FID burner tip. However, a 20 August 1996 e-mail message from a JUMEngineering representative to RTI indicates that a number of their THC instruments are operating atasphalt plants for raw emissions as well as for emissions after treatment.

There appear to be some errors in the draft THC calibration data the MRI submitted for review. Thespan drift data for stack dryer Runs 1 and 2 are identical, which is an unlikely occurrence. Thereappear to be errors in the calculated span drifts for Dryer Stack Run 1 and Load-out Run1. Theseerrors do not appear to be major problems.

I. CAPTURE EFFICIENCY TEST

1. Describe the apparatus forreleasing the tracer gas.

T Tracer gas passes from a compressedgas cylinder through a pressure gaugeand a mass flow meter to four sets oftoggle valves and needle valves, whichare set to deliver 4 L/min. Teflon®tubing goes from each needle valve tosix critical orifices that are set at thebase of one asphalt silo. Each criticalorifice has a maximum flow rate of 0.8L/min.

2. How is the tracer gas measured? T Tracer gas is measured by the extractiveFTIR instrument .

3. How does the tracer gas operatorswitch from one release point toanother release point?

T The tracer gas operator uses the togglevalves to control silo at which tracer gasis being released. Information aboutwhich silo is being used is radioed to theoperator from an observer in the asphaltplant's control room.


COMMENTY N

NA


4. Describe the tracer gas. T Scott Specialty Gases cylinder numberALM013870, 1.99 percent sulfurhexafluoride (SF6) in nitrogen (certifiedworking standard), analyzed on 2 April1998.

5. Does the tracer gas have anappropriate gas mixture andconcentration for the captureefficiency test?

T If the load-out tunnel emissions controlsystem has a flow rate of 15,000 dscm,the FTIR instrument will sample a SF6concentration of 0.19 ppm, which iseasily detected by the extractive FTIRinstrument with a 10-meter pathlength.The SF6 calibration gas has aconcentration of 0.205 ppm , whichclosely matches the sample's SF6concentration

6. What is the analytical uncertaintyof the tracer gas?

T ±5 percent

7. Is the tracer gas traceable to NISTSRMs or otherwise traceable toNIST?

T The tracer gas is traceable to thespecialty gas producer's primarystandards.

8. Has the tracer gas beenindependently verified by MRI?

T

9. Is the tracer gas regulator anddelivery system properlymaintained?

T The equipment appears to have beenproperly maintained.

10. What is the tracer gas flow rate? T 4 L/min

11. Describe how the tracer gas flowrate is measured and controlled.

T The flow rate is measured by a massflow meter and is controlled by a needlevalve.

12. Describe how the tracer gas flowrate meter is calibrated.

T The flow meter was calibrated with aSierra Instruments Cal-Bench™ (serialnumber AN0125), which has a 1%accuracy.

13. When was the last time that thetracer gas flow rate meter wascalibrated?

T 30 April 1998


COMMENTY N

NA


14. Is the tracer gas flow ratecalibration traceable to NISTstandards?

T The Sierra Instruments Cal-Bench™automated primary gas flow calibrationsystem is traceable to NIST length andtime standards.



Midwest Research Institute Standard Data Forms for

Emissions Testing at a Hot Mix Asphalt Plant

Mr. Robert WrightResearch Triangle Institute3040 Comwallis RoadPost Office Box 12 194

SUBJECT TO C/3 ZGiGFIDENTIALBWNESS !NFOHk%TICN CLAIM

Does Not CanfainNafional Security lnfomdion

Research Triangle Park, NC 27709

Dear Mr. Wright,

Enclosed are responses to your request for technical information related to thetechnical system audit of the recent emissions testing at a hot-mix asphalt plantperformed by Midwest Research Institute @4RI). In attempting to respond to your needfor timeliness of a draft report (August 21) all issues easily completed have beenincluded in this response. Several issues, however, cannot be completed at this time, andare expected to be available through our report to EPA which is due by September 30,1998. These issues have been appropriately marked in the following list. I have alsoincluded your original references to “checklist section” for clarity.

1 . (A9) Raw quality control/calibration data sheets and/or calculated calibration data forthe FTlR instrument (e.g. CTS, SF6, or SF61toluene measurements) and the totalhydrocarbon instrument (e.g. calibration error, linearity error, zero and calibrationdrift, sampling system bias.

Both the FTIR and THC met all calibration requirements during the test program,FTIR pathlength determination and CTS calculations are attached. Criteria forpass/fail for the CTS spectra are +/- 5%, which all ofthe CTS spectra met.SF61toluene determinations are not yet complete, but will be included in the finalreport.

THC calibration error, zero drill and calibration drift calculations are also attached.The THC analyzer met all criteria, which include +/- 5% for calibration error, and +/-3% for zero drift and calibration drift. Procedures followed EPA Method 25A, whichrequires calibrations through the sample line. Thus, there are no other linearity orsystem bias checks associated with this method.

Note that the attached data have not been finalized and approved by MRl’s QAprocedures and have been stamped as “Drafi.”

24. (H18) A copy ofthe certificate of analysis for the THC analyzer’s calibration gas andinformation about its analytical uncertainty, traceability, and any independentconcentration verifications.

The THC calibration gas certificate of analysis is attached. The gas is an EPAProtocol gas, and meets all requirements of that certification.

25. (Hl8) Information about the annual factory calibration of the Environics Model 2020gas dilution system and any field evaluations according to EPA Test Method 205.

Calibration records for the Environics gas dilution system are attached

26. (H30) Information about any maintenance logs for the THC analyzer

THC maintenance is performed on an as needed basis in MRI’s laboratories and inthe field. No other maintenance logs are available.

I hope you find the attached information helpful and allow you to complete thosesections of your audit. If you have additional questions, please feel free to call me at816-753-7600, ext. 1228.

Sincerely,

Scott W. KlammEnvironmental Engineer

cc: J. HosenfeldJ. BalsingerT. GeyerM. Toney

2 . (AIO) Blank copies of data sheets used for recording load-out operations data andtracer gas release data.

Blank copies of data sheets used for recording load-out operations data and tracer gasrelease data are attached.

3 . (A14) Identification ofany MRI personnel performing reviews of QC data from theemissions testing.

MRI has a formal review process for all reports generated under this contract. Thereview process includes a review and signature requirement by the Work AssignmentLeader, Program Manager, and Senior Quality Assurance Officer. Specific to thisproject, these individuals would be Mr. Scott Klamm (Field Team Leader), Dr. TomGeyer (FTIR Oversight), Mr. John Hosenfeld (Work Assignment Leader), Dr. BruceDiel (OPPT Program Manager), and Mr. Jack Balsinger (Senior Quality AssuranceOflicer). These individuals are in accordance with the organizational chart (Figurel-l) ofthe QAPP.

4 . (C3) Information about any current summaries (e.g. training files) of the training andqualifications of each project team member.

MRI maintains a file of personnel resumes and curriculum vitae for each project teammember detailing their experience and qualifications. Single page summaries foreach team member are attached.

5. (C4) Information about any descriptions of individual project team memberresponsibilities.

Individual project team member responsibilities are briefly summarized in the QAPP,Section 1.4. Specific responsibilities not described in the QAPP were assigned atMRI project coordination meetings prior to the field phase, and were updated on adaily basis during the field effort

6 . (D4) Information about the source (e.g. documented performance criteria or actualQC data) for the data quality indicator goals for the sample concentration FTIRmethod.

This comment will be addressed in the report

7 . (F12) Information about noise levels for each FTIR analytical region

As demonstrated to RTI in the field, RMSD noise levels in the sample spectra wereon the order ofO.OOO1 absorbance units in the 1000-1300 wavenumber region, In thereport, RMSD in the residual (subtracted) spectra will also be calculated (i.e. afteranalysis).

8 . (F13) Information about sample absorption pathlengths for each FTIR analyticalregion.

The average pathlength for the test program was determined from the daily CTSspectra, and is included in the CTS stability calculations described earlier in Item 1,and contained in the attachment for Item 1. These preliminary pathlengths werecalculated from raw field parameters and are subject to revision in the report,

9 . (F14) Information about fractional analysis uncertainties for each FTIR analyticalregion.

This comment will be addressed in the report.

10. (F25) A description about how analyte concentrations are quantitated from FTIRabsorption spectra.

These procedures are outlined in EPA Method 320 and the FTIR Protocol included inthe QAPP. Specific aspects of the analyses and deviations from the methods will beincluded in the report.

11. (F26) Information about minimum detectable concentrations for the compoundsfound during the emissions testing by the extractive FTIR method.

A summary of minimum detectable concentrations for non-detects will be included inthe report

12. (F29) A copy of the certificates of analysis for the CTS, the SF6, and the SF6/toluenestandards.

Copies of certificates of analysis for the CTS, SF6, and SF6itoiuene sases have beenattached.

13. (F30) Information relating to the basis for selection ofthe CTS gas mixture andconcentration relative to the selection criteria given in Section 4.5 ofthe FTIRProtocol.

Ethylene has been used as a CTS gas on numerous EPA test programs, and meets thecriteria given by the FTIR Protocol.

14. (F3 1) Information about the uncertainty of the analysis of the CTS, the SF6, and theSF6/toluene standards.

Analytical accuracy for each of these gas standards is contained on their respectivecertificates of analyses. All are listed as +/- 5%.

15. (F32) Information about the traceability ofthe analysis ofthe CTS, the SF6, and theSF6Itoluene standards to NIST.

Scott Specialty Gases uses gravimetric procedures to generate their gas standards.They do not claim NIST traceability, but, rather, are certified by their own internalcalibrations and standard procedures.

16. (F33) Information about any independent concentration verifications of the CTS, theSF6, and the SF6koluene standards by h4RI.

See Items 14 and 15, above

17. (F34) Information about the variability of successive CTS measurements.

Variability of successive CTS measurements will be included in the report, and canbe seen from the calculations presented earlier in Item 1. The CTS measurementsmet the Protocol requirements of +/- 5% precision.

18. (F37) Information about any maintenance logs for the FTIP. instrument

CTS stability is the primary indicator of instrument operations. When n e c e s s a r y ,FTIR maintenance is performed on an as needed basis in MRI’s laboratories and inthe field. For this particular test program, the instrument was serviced by an Analectrepresentative on July 20, was found to be in good working order, and did not requireany corrective actions.

19. (G3) Information about any Entropy Environmentalists or MRI laboratory or fieldvalidations of the sample concentration FTIR method for the compounds foundduring the emissions testing.

This information was provided to RTI while in the field (Fourier Transform InfraredMethod Validation at a Coal-tired Boiler, Entropy Environmentalists, July 1993,published by EPA in 1994).

20. (G12) Information about the calibration traceability of the wet test meter used tocalibrate dry gas meters in the VOST sampling trains associated with the sampleconcentration FTIR method.

MRI calibrates the dry gas meters versus a wet test meter located at our Kansas Citylaboratories, The wet test meter is calibrated according to the displacement methodand is, therefore, considered a primary standard. Calibration records for the wet testmeter and the VOST consoles are kept on tile and are available upon request.

21. (G19) A copy of the certificate of analysis for the toluene-d8 surrogate spiking gasand information about its analytical uncertainty, traceability, and any independentconcentration verifications.

A copy of the toluene-d8 certificate of analysis is attached.5%.

The gas is certified to +/-

22. (G3 1) Information about any Entropy Environmentalists or MRI laboratory and/orfield determinations of the collection and desorption efliciencies of the sampleconcentration FTIR method for the compounds found during the emissions testing.

See Item 19. above.

23. (G32) Information about minimum detectable concentrations for the compoundsfound during the emissions testing by the sample concentration FTIR method.

Minimum detectable concentrations for this method are, by nature, dependent uponthe compound and sampling parameters, and will be estimated in the report. For thistest, sampling parameters were chosen to obtain 100 ppb of toluene and xylene,which were identified by PES in a preliminary screening.

c

Attachment 1

Item 1 (A9)

Filename ctspath Temp. (C) Pathlength (m) % Diff.

co721 b 2.7207 124 10.812 3.22~0721~ 2.6321 124 10.460 -0.14c072Id 2.6718 124 10.618 1.37c072Ie 2.6181 124 10.404 -0.67c0722a 2.6234 124 10.425 -0.47c0722b 2.6280 124 10.444 -0.30c0723.a 2.6131 124 10.364 -0.86c0723b 2.6162 124 10.397 -0.74~0723~ 2.6420 124 10.499 0.24c0724a 2.6242 124 10.429 -0.44c0724b 2.6536 124 10.545 0.68~0725~1 2.6122 124 10.381 -0.89c0725b 2.6247 124 10.430 -0.42~0725~ 2.6582 124 10.564 0.85c0726a 2.6390 124 IO.487 0.12c0726b 2.6378 124 10.483 0.08cO727a 2.6220 124 10.420 -0.52c0727b 2.6268 124 10.439 -0.34~0727~ 2.6283 124 10.445 -0.28c0727d 2.6234 124 10.425 -0.47

CTS Pathlength and % Difference Calculations

Average = 10.475

DRAFT

Pathlength is based on use of a 99.9 ppm ethylene standard.% Difference is based on the average calculated pathlength.

THC

Cal Gas Measured Difference

Value Value As % Error

0.0 0.0 0.0 Pass90.3 90.2 0.1 Pass50.2 50.9 1.4 Pass25.0 25.0 0.0 Pass

Run 1 - Dryer Stack - 7121198

Calibration Error Determination

Pass/ Fail

Instrument Span for THC is 100 ppm

Pass/Fail Criteria is +I- 5% of Cal Gas for THC

Zero Drift

Initial 1 st Drift Check Difference


0.0

1 st Drift Check Final DifferenceValue Value As % Error

0.3

Initial 1 st Drift Check DifferenceValue Value As % Error

90.2

1 st Drift Check Final DifferenceValue Value As % Error

0.3 0.3

0.3 0.0

Span Drift

89.9 0.7 Pass

Pass/Fail

Pass

Pass/Fail

Pass

Pass/Fail

Pass/Fail

89.9 89.5 0.4 Pass

Run 2 -Dryer Stack - 7122198 DRAFT


Cal Gas

Value

Measured

Value

Difference

As % Error

Pass/ Fail

0.0

90.350.2

25.0

0.2 0.2 Pass90.1 0.2 Pass50.6 0.8 Pass25.6 2.4 Pass

THC


Pass/Fail Criteria is +/- 5% of Cal Gas for THC

Zero Drift

1 st Drift CheckValue

Difference

As % Error

Pass/Fail

Value

0.2 -0.1 0.3 Pass

Pass/Fail1 st Drift CheckValue

-0.1

FinalValue

DifferenceAs % Error

0.3 0.4 Pass

Span Drift

initialValue



Pass/Fail

90.2 89.9 0.7 Pass

Pass/Fail1 st Drift Check FinalValue Value


89.5 0.4 Pass89.9

. .

Run 1 - Load Out - 7/23/98 (aborted)


Cal GasValue

MeasuredValue


Pass/ Fail

THC Silo 0.0 3.6899.0 900.0498.0 504.0249.0 256.0

Instrument Span for THC Silo is 1000 ppm


0.40 .01.22 .8

PassPassPassPass

Cal Gas Measured DifferenceValue Value As % Error

Pass/ Fail

THC Tunnel 0.0 0.290.4 90.250.2 49.825.0 27.7

Instrument Span forTHC Tunnel is 100 ppm


0.2 Pass0.2 Pass0.8 Pass1.2 Pass

Zero Drift

Initial Final DifferenceValue Value As % Error

THC Silo 0.0 9.8 1.0

Initial Final DifferenceValue Value As % Error

THC Tunnel 0.0 0.5

Initial FinalValue Value

0.5

Span Drift

Difference

As % Error

THC Silo 900.0 914.0 1.4

Initial Final Difference


THC Tunnel 90.2 89.7 0.5

Pas&Fail

Pass

Pass/Fail

Pass

Pass/Fail

Pass

Pass/Fail

Pass

Pass/Fail Crtteria for Drift is +I-3% of THC Span

.

Run 1 - Load Out - 7124198


Cal Gas MeasuredValue Value

THC 0.0 1.2899.0 905.0498.0 508.0249.0 246.0

instrument Span for THC Silo is 1000 ppm


THC


1.2 -0.2


THC Silo 905.0 907.0

Pass/Fail Criteria for Drift is +/-3% of THC Soan


Pass/ Fail

0.10.72.01.2

PassPassPassPass

Zero Drift


0.1

Span Drift


0.0

Pass/Fail

Pass

Pass/Fail

Pass

. .

Run 2 - Load Out - 7/25/98


Cal GasValue

Measured DifferenceValue As % Error

THC 0.0 1.7899.0 902.0498.0 506.0249.0 254.0


Pass/Fail Criteria is +I- 5% of Cal Gas for THC

0.2 Pass0.3 Pass1.6 Pass2.0 Pass

I-HC

InitialValue

1.7

FinalValue

2.0


THC Silo 902.0 900.0

Pass/Fail Criteria for Drift is +/-3% of THC Span

Zero Drift


0.0

Span Drift


0.0

Pass/ Fail

Pass/Fail

Pass

Pass/Fail

Pass

.

lntermittant Load Dump - 7125198


Calm Gas Measured DifferenceValue Value As % Error

THC 0.0 0.290.4 90.7

50.2 50.9

25.0 25.7Instrument Span for THC is 100 ppm


0.20.31.42.8

THC


0.2 0.4


THC Silo 90.7 90.4

Pass/Fail Criteria for Drift is +/-3% of THC Soan

Zero Drift


0.2

Span Drift


0.3

Pass/ Fail

PassPassPassPass

Pass/Fail

Pass

Pass/Fail

Pass

Run 4 - Baseline - 7126198


Cal Gas

ValueM e a s u r e d Difference

Value As % ErrorPass/ Fail

THC 0.0 0.2 0.2 Pass90.3 90.5 0.2 Pass50.2 51.1 1.8 Pass25.0 25.5 2.0 Pass



Zero Drift

InitialValue



Pass/Fail

0.2 0.1 0.1 Pass

Pass/Fail1st Drift Check FinalValue Value


0.1 0 .1 0.0 Pass

Span Drift

InitialValue

1st Drift CheckValue


Pass/Fail

90.5 90.4 0.3 Pass

Pass/Fail1st Drift Check FinalValue Value


90.4 90.8 0.4 Pass

I .

Run 3 - Load Out - 7123198



THC Silo 0.0 1.1899.0 907.0498.0 504.0249.0 260.0

Instrument Span for THC Silo is 1000 ppm


0 .10 .91.24.0


THC Tunnel 0.0 0.290.4 90.450.2 50.925.0 25.5

Instrument Span for THC Tunnel is 100 ppm


0.20.01.32.0


THC Silo 1.1 -0.1


THC Tunnel 0.2 0.1


THC Silo 907.0 903.0


THC Tunnel 90.4 90.6

Pass/Fail Criteria for Drift is +/-34/o of ~7tiC Span

Pass/ Fail

PassPassPassPass

Zero Drift


0.1


0.1

Span Drift

Difference

As % Error

0.0

Difference

As % Error

0.2

Pass/ Fail

PassPassPassPass

Pass/Fail

Pass

Pass/Fail

Pass

Pass/Fail

Pass

Pass/Fail

Pass

. .

Analyzer

THC Silo

THC Tunnel

THC Dryer Stack

Response Times DRAFT

Response Time

1 min. 25 sec.

35 sec.

1 min. 30 sec.

Attachment 2

Item 2 (AlO)

.

Load-Out Log Spreadsheet Sheet __ of

. .

SF6 Gas Delivery data Spreadsheet Sheet-of-

Attachment 3

Item 4 (C3)

Thomas J. Geyer

Principal Chemist

Dr. Geyer specializes in molecular spectroscopy, FTIR test method development andvalidation, EPA protocol development, FTIR data analysis, analytical softwaredevelopment, and project management. Dr. Geyer has developed EPA FTIR applicationsfor emissions testing in the field and managed the first field evaluation of an FTIRcontinuous emission monitoring system for measuring HAPS. He has led many field testprograms and developed analyte spiking procedures for hazardous air pollutants (HAPS).Dr. Geyer helped develop the EPA FTIR analytical protocol to measure reference spectraand to apply spectra for field measurements.

Recently, Dr. Geyer developed FTIR test Methods 3 18 for wool fiberglass, 320 forhazardous air pollutants and Performance Specification 15 for FTTR continuous emissionsmonitoring applications. He also does analytical programming for instrumental methods.Based at MRI’s North Carolina Office, Dr. Geyer develops hardware, software, andmeasurement techniques using FTIR spectrometry for the analysis of pollutant emissions.His project assignments have included a number of FTIR method development and testprograms.

Before joining MRI in 1995, Dr. Geyer was a Senior Chemist and Group Leader in theResearch Division of Entropy, Inc. He directed an EPA FTIR emission test project at fiveelectric utilities, conducted field studies with FTlR emissions tests, and managed in-houselaboratory projects. He helped develop procedures to analyze FTIR validation data andhelped develop FTIR validation sampling procedures. He directed laboratory and fieldevaluations to exp,and the capabilities ofFTlR emissions testing.

From 1989 to 1991, Dr. Geyer was Assistant Professor of Chemistry at the U.S. NavalAcademy. Previously, he spent a year as an Offme of Naval Technology PostdoctoralFelllow at China Lake, California, He performed studies to characterize material ejected bylaser ablation from the surface of YBa,Cu,O, high-temperature superconductor. As aResearch Assistant at the University of South Carolina from 1984 to 1988, he conductedoriginal research in molecular spectroscopy to determine structures and conformationalequilibria of substituted cyclobutane and cyclopentane molecules.

Dr. Geyer has a Ph.D. in Physical Chemistry from the University of South Carolina(1988). He is a member of the American Chemical Society, American Physical Society, Airand Waste Management Association, and Sigma Xi (Scientific Research Society of NorthAmerica). Entropy honored him with an Outstanding Contribution Award in 1992. He isthe coauthor of more than 20 technical publications and presentations. l~)Wr

Scott W. Klamm

Senior Environmental Engineer

Mr. Klamm specializes in air toxics, combustion processes, emission controltechnologies, and related environmental engineering programs. He has worked extensivelyon the development and operation of laboratory and field equipment for point source andambient testing. His recent activities at MRI include leading an EPA field methodcomparison project on acetaldehyde, performing data analysis for a comparison of threemethods for HI?, and leading the data analysis and reporting activities for an EPAincineration project to identify and measure products of incomplete combustion using awide range of field and analytical methods. Mr. Klamm has also recently led field studiesfor two industrial clients to demonstrate collection efficiency on a control device in one, andto determine measurement errors of an on-line mass spectrometer for emission monitoringin the other.

Mr. Klamm developed an experimental extended-period ambient organic sampler andcarried out IaboratoIy testing of the system. For the electronics industry, he examined theuse of CFC solvents and alternative technologies for emission control. Mr. Klammprepared the test plan and matrix design as well as special sampling equipment for a fieldstudy of organic compound adsorption in boiler soot for EPA. On several occasions, he hasserved as an EPA field auditor for the disposal of chemical agents and munitions at an Armyfacility. He ran tests on stack sampling equipment and emissions monitors and maderecommendations on the equipment and monitors. Mr. Klamm has also been field auditorand reviewer of trial bum data as part of permit review assistance to the states of Utah andKansas, EPA Headquarters, and EPA Region VII.

As part of an evaluation of the Alaska CO inventory, Mr. Klamm tested variouslocations in an urban area for mobile-source CO emissions, applied meteorologicalinformation, assisted in modeling, determined accuracy of existing monitors compared toMRI-gathered data, and helped prepare recommendations to the State of Alaska. For theU.S. Air Force, Mr. Klamm designed a test matrix to study airborne pollutants and toxicgases formed during extinguishment of jet fuel fires with Halon agents at firefighter trainingexercises. He assisted in operation of SuMMq ambient VOST, open-path FTIR, PUF,particulate, and CO monitors. He also performed data reduction, coauthored reports andpapers, and assisted in risk assessment and dispersion modeling,

Regulatory work has involved coauthoring EPA guidance documents on setting permitconditions, reporting trial bum results, and related incineration topics, such as combustiongas velocity, treatment of gasoline-contaminated soils, and solid waste feed systems.

Mr. Klamm graduated magna cum laude from the University of Missouri-Rolla with aB.S. in Chemical Engineering in 1985. He has taken additional courses in solid andhazardous waste management, human environmental biology, and dispersion modeling. Heis a member of the Air and Waste Management Association. He received the 1990 BidstrupAward from MRI for outstanding technical contribution and the I997 Achievement Awardfrom MRl’s Council of Principal Scientists. 1~0j2r j1’>8

. .

Robert G. Gulick

Supervisor of Field Programs

Mr. Gulick specializes in the operation and maintenance of continuous emissionmonitors in compliance with appropriate EPA methods. As Supervisor of FieldPrograms, Mr. Gulick is responsible for assuring that all equipment for source samplingis available, calibrated, and in working condition. In addition, he has responsibility foroperating and maintaining continuous emission monitors during field sampling atindusnial facilities. Recent programs include:

l Performed continuous emission monitoring of carbon monoxide, carbon dioxide,sulfur dioxide, nitrogen oxides, total hydrocarbons, and hydrogen chloride onsource sampling projects. These have included a hysteresis boiler study inPennsylvania and a kiln study in Nebraska.

l Evahated 24-hour sampling for total hydrocarbons to determine captureefficiency in the development of total temporary enclosure model.

l Project Leader on a shldy that assessed and recommended a full complement ofcontinuous emission monitors for a hazardous waste facility’s continuousemission monitor replacements.

In 1988 and 1989, Mr. &lick was a Technical Sales Engineer with TorotelProducts of Grandview, Missouri. He negotiated price and delivery of electroniccomponents, expedited production to meet delivery schedules, and monitoredengineering changes to meet customer specifications. During 1986 and 1987,Mr. Gulick was the owner and manager of a home electronics retail sales and servicebusiness. He was responsible for retail sales, inventory purchase and control, serviceand repair of audiovisual and computer products, advertising, and personnelmanagement.

Mr. Gulick’s background includes rive years as a Field Service Engineer withPerkin-Elmer. He installed and repaired analog and digital test instrumentation,computers, and peripherals. He handled inventory control, trained junior engineers, andsold service contracts.

In 1981, Mr. Gulick received an A.A.S. in Electronic Engineering Technology fromWestern Iowa Technical College. In 1990, he completed special training in hazardouswaste operations. He has certification of OSHA safety and health training.

A-16

,

Andrew E. Page

Assistant Environmental Scientist

Mr. Page performs work in the Air Quality Engineering Section of the AppliedEngineering Department. His experience includes field and laboratory testing for fugitivedust sources as well as ducted emission sources using various types of sampling equipment.He performs laboratory analysis on emission sources; Method 18 analysis on ductedemission sources by gas chromatography, including data reduction for these analyses; andassists in operation and maintenance of mobile laboratory and field sampling equipment andanalytical instrumentation (JR and GC with FID, PID, and ECD).

From 1994 to March of 1997, Mr. Page worked for Ken Wilcox Associates where hetested leak detection systems and prepared certification reports according to EPAprotocols. While at KWA he gamed extensive experience in working with and testing leakdetection equipment for aboveground and underground storage tanks and pipeline systems.His other responsibilities included assistance in development of a fuels management researchcenter; product development testing for clients; data reduction and analysis for varioustypes of testing; preparation of estimates for analysis on environmental cleanup projects;soil and water analysis using gas chromatography (GC) and infrared spectrometry (IR); andmaintenance of GC and & calibration, troubleshooting, and repair.

As a research and teaching assistant at the University of Missouri, from 1992 to 1994,he worked on various research projects, namely development of reverse-bum gasification asa thermal treatment process for petroleum sludge, sewage sludge, and other wastes.Presented research at various conferences. As a teaching assistant, he taught help sessionsand laboratory sessions for general chemistry courses.

Mr. Page received a B.A. in Chemistry from Central Methodist College in 1993

.

Pamela S. Murowchick

Associate Environmental Engineer

Ms. Murowchick provides engineering support on environmental contracts rangingfrom incinerator trial bums to protocol development to test monitoring systems. Thediversity of her assignments has required versatility in applying her engineering knowledgeand the ability to work effectively with project management, project team members, andclient personnel. Her experience at MN has included:

* Prepared trial bum plans and associated QA plans.* Operated the sampling console for Modified Method 5 testing during RCRA and

TSCA trial bums of hazardous waste incinerators.* Performed laboratory preparation and recovery tasks for incinerator sampling and

analysis projects.* Performed test data reduction and reporting.* Assisted on EPA regulatory support assignments involving waste management.* Coauthored a conference presentation on incineration options for waste

management.* Coauthored a conference presentation on emissions of organic compounds and

combustion gases from an industrial boiler during hazardous waste cofiring.* Assisted with EPA protocol development work for underground storage tanks at

MRI’s Experimental Tank Facility.* Assisted with evaluations of commercial equipment for tank tightness testing.* Assisted on a study of solar energy for waste disposal.

From 1988 to 1989, Ms. Murowchick was a Plant Chemist with Lubripac in KansasCity, Kansas. She performed routine physical and chemical testing of oil in the QualityControl Laboratory. Ms. Murowchick monitored the progress of special projects involvingcleanup in the tank farm, and she tested and adjusted new oil formulations in the laboratorybefore full-scale production was begun.

Ms. Murowchick received a B.S. (with honors and distinction) in Chemical Engineeringfrom Pennsylvania State University in 1985. In 1987 and 1988 she did course work inSecondary Education and Chemistry at Mercyhurst College. She is certified in CPR andcompleted OSHA’s RCRA safety training. Her computer skills include Fortranprogramming and WordPerfect, Quattro, and Lotus applications.

.

James S. Surman, Jr.

Senior Environmental Scientist

Mr. Surman has over 28 years of direct experience in stationary source air emissionstesting, continuous emission monitoring, and related QA/QC procedures. His work at MRIfocuses on the evaluation of hazardous waste incinerators, municipal waste combustors,boilers and industrial furnaces, and other sources of air pollution. He has been FieldSampling Task Leader on 63 projects and Project Leader on 8 emission studies ofincinerators, boilers, and kilns, Mr. Surman has also participated as an operator andsampler on numerous field programs and has audited RCRA trial bums of hazardous wasteincinerators and emissions tests at power plants for regulatory agencies. In addition to fieldassignments, he participates in EPA-sponsored work assignments relevant to qualityassurance and emissions standards development.

Field assignments have involved sampling for volatile and semivolatile organiccompounds; particulate matter, including PM,, by the EGR method; particle size distri-bution; multiple metals, including hexavalent chromium, acid gases, and other emissions.Mr. Surman has prepared and assisted engineers in preparing proposals, test plans, workplans, QA plans, and project final reports. He also has been responsible for the hands-onpreparation and execution of field sampling projects involving as many as 15 field personnelfor periods up to one month at a test site.

From 1978 to 1987, Mr. Surman was Quality Assurance Manager and ProjectSupervisor with Mostardi-Platt Associates of Bensenville, Illinois. He evaluated all testingprocedures, equipment maintenance and calibrations, and techniques in the performance ofnew or innovative test methods. He helped implement EPA reference methods as theybecame available and supervised air pollution source testing and air pollution controlevaluation for various industries. He also conducted ambient air monitoring studies andaudits and developed a QA plan for testing activities.

Earlier experience with air pollution source testing was with Kin Associates at ChicagoHeights, Illinois (1977) and Commercial Testing and Engineering Company in SouthHolland, Illinois (1973-1977). Mr. Surman initially worked in Commercial Testing andEngineering Company’s services for the sampling and analysis of coal, iron ore, water, soils,and other materials for chemical and physical characteristics (1970-1973). With RandDevelopment Corporation in Cleveland, Ohio (1966- 1970) he aided in field investigationsrelated to water pollution control, groundwater contamination, and strip mine reclamation.

Mr. Surman received the B.S. in Biology with a minor in Chemistry horn ClevelandState University (1966). He is a member ofthe Source Evaluation Society and coauthor ofa municipal waste combustion multipollutant study (EPA-600/8-89.064).

Bob J. Edwards

Senior Technician

Mr. Edwards joined MRI’s Field Measurements Section in 1992. He has participated ina number of industrial projects around the country for hazardous waste incinerator testingand analysis services. He also has worked on assignments under MRI’s subcontract fortechnical assistance to EPA for the implementation of RCRA regulations for hazardouswaste management facilities.In support of various field sampling and analysis programs, Mr. Edwards calibratesequipment, performs routine and corrective maintenance on equipment, operates an EPAMethod S console during stack sampling, performs Orsat analysis on gas streams, andtitrates EPA Method 6 samples. He stages and destages equipment for field sampling andsets up the equipment on-site. In addition, he performs data entry for reports, investigatesequipment and vendors to assist with equipment purchasing decisions, and orders materials.

His background includes six years in the U.S. Navy, followed by three years inmanufacturing and industrial facilities. He has worked extensively in environmentalmaintenance, pneumatics, hydraulics, and electrical and mechanical equipment and systems.Mr. Edwards also has experience with quality assurance, radiological controls, gaugecalibration, hazardous materials, and security and safety issues.

In the Navy, Mr. Edwards was a missile technician. He operated and maintained thePoseidon Ballistic Missile System as launcher operations supervisor. He performedcolateral duties, such as nuclear weapons handling supervisor, departmental coordinator formaintenance and material management, nuclear weapons radiological controls assistant,missile and readiness equipment expert, departmental publications coordinator, qualityassurance inspector, field metrology technician, and nuclear weapons security guard. Hecompleted several training programs in the Navy that covered electricity, electronics,hydraulics, pneumatics, digital computers, radiological controls, QA, gauge calibration,theory and practice of equipment maintenance, and related specific disciplines. He has alsotaken courses in electronic engineering at Maple Woods Community College in KansasCity.

With a division of Kraft-General Foods, Mr. Edwards worked as plant maintenancesupervisor/trainee. He oversaw maintenance aspects of production and packaging,interacted with contractors and the company upgrade team in monitoring progress duringequipment improvement and expansion, and served on the plant’s hazardous materialsresponse team after required OSHA training.

2%

Daniel 0. Neal

Senior Technician

Mr. Neal provides technical support on programs for environmental monitoring,sampling, and analysis in the Field Measurements Section of MRPs EnvironmentalTechnology and Engineering Department. His responsibilities include staging and destagingequipment for field tests, setup and operation of equipment, and calibration andmaintenance of test equipment.

Since joining MRI in 1993, Mr. Neal has accompanied field measurement teams onseveral projects for clients in government and industry. In addition to equipment setup, heperforms various tasks, including nmning the Method 5 console and operating the stackprobe. Project work has included an emissions test of a kiln at a brick manufacturing plantfor EPA, compliance. and performance tests of cooling towers, and various trial bum,minibum, and performance tests for industrial clients.

Prior to joining MET, Mr. Neal worked as a sales representative for a publishingcompany and an appliance company. From 1976 to 1984, he worked for the PowerDivision ofBums and McDonneh Engineers in Kansas City, Missouri. As a Design Detailerand Electrical Substation Drafting Supervisor, Mr. Neal assisted with design and layout ofpower substations, Mr. Neal has completed undergraduate courses at Kansas StateUniversity, Johnson County Community College, and Longview College.

Attachment 4

Items 12,21, and 24 (F29, G19, and H18)

~scott Specialty Gases

,.,-.&wed6141 EASTON ROAD PO BOX 310PLUMSTE~VILLE ?A 1E949-0310

BFrOm : Phone: 215-766-8861 -Fax: 215-766-2070

C E R T I F I C A T E O F A N A L Y S I S-------------______------------------------------------------------------MIDWEST RESEARCH PROJECT #: 01-01788-004SCOTT KLAMM PO#: 033452425 VOLKER BLVD ITEM #: 01023822 5AL

DATE: 4/07/98KANSAS CITY MO 64110

-----_-_--_--__----------------------------------------------------------CYLINDER #: ALM033887 ANALYTICAL ACCURACY: +/-5%FILL PRESSURE: 2 0 0 0 PSIG

BLEND TYPE : CERTIFIED WORKING STDREOUESTED GAS

COMPONENTSULFUR HEXAFLUORIDENITROGEN

mmLYSIS- ~~.~CONC MOLES (MOLES)

.2 P P M 0 . 2 0 5 PPMBALANCE BALANCE

Scott Specialty Gases6141 EASTON ROAD PO BOX 310

a

Shipped PLUMSTEADVILLE PA 18949-0310From: Phone: 215-766-8861 Fax: 215-766-2070

C E R T I F I C A T E O F A N A L Y S I S----------------_---____________________--~----------~-------------------MIDWEST RESEARCHSCOTT KLAMM425 VOLKER BLVD

KANSAS CITY

PROJECT #: 01-01788-001PO#: 033452ITEM #: 0102S3000815ALDATE: 4/07/98

MO 64110

---_------_-____--_-____________________---------------------------------CYLINDER #: AALFILL PRESSURE:

ANALYTICAL ACCURACY: +/-5%1280 PSIG

BLEND TYPE : CERTIFIED WORKING STDREQUESTED GAS ANALYSIS

COMPONENT CONC MOLESSULFUR HEXAFLUORIDE

(MOLES)4 . PI-9 3.83 P P M

TOLUENE 100. PEW 105. PPMNITROGEN BALANCE BALANCE

.

i0 N o Xl3452

___--..__ <IL- _-____ -______ _-- _____________________________Reordcr/Serv,ce Contact ~21517s6-8a61 FttNSTER!xlLLE PR ISrn

CEFtTlFlCATE OF ANALYSIS

06/10/96 15:29 axz5 766 2070 3b"ll Jlz.cAnr--. . _- i

-:

@

Scott Specialty Gases6141 BASTON ROAD PO BOX 310

Shipped PLUMSTEADVILLE PA 18949-0310From: Phone: 215-766-8861 Fw: 215-766-2070

CERTIFICA.TE O F A N A L Y S I S_____-_________--_------------------------------------------------..------MIDWEST RESEARCH PROJECT #: 01-017118-005SCOTT IcLAm PO#: 033452425 VOLKER BLVD ITEM #: 01021951 5AL

DATE: 3/31/98KANSAS CITY M O 64110

----____-_--__-_____----------------------------------------------. ------

CYLINDER #: AU4005893 ANALYTICAL ACCURACY: +/-StFILL PRESSURE: 2000 PSIG

BLEND TYPE : CERTIFIED WORKING STDREQUESTED GAS ANALYSIS

COMTONENT CONC NOLES (MOLES)100. PPM 99.9 PPM

NITROGEZN E3sANcE BAL:.NcE

~ I'i.RFi~ili~lE!) I~CCOROIIIC 'I'0 EPA 'TRACEABILITY I'IOTOCOL FOR ASSAY AND CERTIFICATION OF GASEOUS CALIBRATION STANDARDS (PRoCED""E #till<Al stomer : Order No:! itrn PRODUCTS A,,,> CfIL:,,ICALS, 1,rc.

312-020638-01Batch No:

i 1 R CAI.!DE,I STREE:T861-33582

Notes:

Attachment 5

Item 25 (H18)

Daniel A. KaplinskiSales Engineer I I

vironics Inc. uo,,,d. CT ‘3608-ltij Industrial Park Road 6%(660) 8,2 -111, . FAX: (860) 8i’o-9333World Wide Web: http:llwwW.OnVirOniCs.cornE-mail: dktaplinski8environics.com

COmpU,,nl~ Ga, ~i,;~p~oi~ytion~c~~~~~~~~ sw*m*

Mf #: I<

ENVIK@NICS FLOW CONTRCLLEF: C&LIERATION SHEET

D e s c r i p t i o n : A I R ? S i r e : 1 rw @iB . SCCM. tr:-f a c t o r : 1.!3

This f l a w enntroller was c a l i b r a t e d clsing a S i e r r a C a l Eench(TM)I a t r a c e a b l eP r i m a r y F l o w S t a n d a r d C a l i b r a t i o n S y s t e m . T h i s c a l i b r a t i o n i s r e f e r e n c e d t od r y a i r at. a temperaturr a f LF (-C) a n d a p r e s s u r e nf 29.92 in.Hg (76BTorr).

Mf #: 3, Description: AII? 5i-E.: -. 1 gg [.:I . !!I Sl3Ctq ~ K-factor: 1. i?i

SERIAL # nw 9572K1

This flow controller was calibrated using a SiF-t-r.5 Cal i&-,~;?<Tt'l! ~ a traceablePrimary Flow Stcindard Ca!ibration~Systc,n. This calibration is referenced todry air at S tetTPC+-ature nf u I-C) and a pressure zf :?.'?I1 in.Hg [76WTo~r).

T h i s f l o w cantrnller w a s c a l i b r a t e d u s i n g a S i e r r a Cal F.enrI-, (TM) ~ a trace&SEPr imary F low Standard Ca l ib ra t ion System. T h i s calihratinn i s refcrencd t odry sir a t s t e m p e r a t u r e o f RF (-C! a n d a pr~?~surp nf 27.?2 in.Hq (760Tar.ri.

This

is

a

Blank

Page

Chapter 3

Technical Systems Audit

Manual Emissions Test Methods Performed by

Pacific Environmental Services Inc.

at Hot Mix Asphalt Plant C,

Los Angeles, CA

This

is

a

Blank

Page

3.1Technical Systms Audit of Test Methods Performed by PES

3.0 TECHNICAL SYSTEMS AUDIT - PACIFIC ENVIRONMENTAL SERVICES INC.

This Chapter is the final technical systems audit (TSA) of the manual emissions testing at HotMix Asphalt Plant C in the Los Angeles, California area performed by Pacific Environmental Services,Inc. (PES). As explained in detail in Chapter 1, the final TSA includes revisions by the EPA QualityAssurance Officer. The draft TSA report from Research Triangle Institute (RTI) was delivered to theEPA Quality Assurance Officer on September 29, 1998. The TSA was performed by R.K.M. Jayantyand Robert S. Wright of Research Triangle Institute (RTI) under EPA Contract 68-D4-0091, workassignment 99-03, from July 20, through July 26, 1998. The TSA was conducted in accordance withprinciples described in the EPA Quality Assurance Division’s working draft version of EPA Guidancefor Technical Assessments for Environmental Data Operations (EPA QA/G-7).

In general, PES and its team members have performed the testing according to the proceduresoutlined in the Site Specific Test Plan (SSTP) and Quality Assurance Project Plan (QAPP). Anydeviations from the QAPP or SSTP have been discussed with the EPA Work Assignment Manager. PESand its team members who are present at the site are well qualified and experienced in the work to whichthey are assigned.

Item Findings that may have a potential effect on data quality

1 The EPA Methods, VOST and Modified Method 5 (MM5), have not been validatedfor all the chemical compounds of interest in the asphalt plant emissions. Some ofthe compounds may have been recovered during the laboratory testing in otherprograms, but none have been validated in the asphalt plant emissions matrices. This was a research study, however, and not meant to have validation.

2 The VOST cartridges were spiked with surrogate compound(s) before fieldsampling at Triangle Labs.

3 The precision and accuracy goals indicated in the QAPP are based on estimatesfrom the VOST and semi-VOST methods, but they are not established values formany of the compounds of interest in this program.

Item Findings that are unlikely to have an effect on data quality

1 The spikes into the XAD-2 resin (MM5) and Anasorb 747 (M-18) were loadeddirectly onto the cartridges in the field.

2 The train operators have noted the readings on the data sheets. The data sheets donot have any column to note any comments during sampling. Similarly, a notebookwas maintained at the site to record any problems or process changes, etc., duringsampling.

3 The flow rates used for VOST at the tunnel exhaust were too low to maintainconstant on the rotameter or VOST console, which may cause some error in the totalvolume collected.


Technical Systems Audit Checklist

Project: Emissions Testing at a Hot Mix Asphalt Plant Organization: Pacific Environmental Services Location: Asphalt Plant C - Los Angeles, CA Assessors: R.K.M. Jayanty and R.S. Wright (Research Triangle Institute; Research Triangle Park,

North Carolina) Audit Dates: July 20 through July 26, 1998 Brief Project Description: EPA is investigating hot mix asphalt plants to identify and quantifyparticulate matter and hazardous air pollutants (HAPs) emitted from asphalt cement load-out operations. EPA has issued a work assignment to PES to conduct an air emissions test program to collect data insupport of the investigation. Asphalt Plant C in Los Angeles, California was chosen primarily becauseload-out emissions are controlled by a silo exhaust system and a load-out tunnel. The plant has aproduction capacity of more 650 tons per day. Approximately 2,000 tons per 4 hour period wereproduced, during the test. The primary objective of the project will be to characterize air emissions oforganic HAPs from asphalt cement load-out operations and operation of the hot mix dryer; however, PESis not responsible for the testing of the hot mix dryer stack. Testing will be performed to characterizeemissions from the storage silos, the load-out tunnel, and the hot mix dryer.


COMMENTY N

NA

A. GENERAL QUALITY ASSURANCE INFORMATION

1. Is there an approved qualityassurance (QA) plan for the overallproject and has it been reviewed byall appropriate personnel?

T A QAPP was submitted to EPA inJanuary 1998. Subsequently, tworevisions have been made and a finalQAPP was submitted on July 14, 1998.Additionally, PES submitted a SSTP inJune 1998. Both documents have beenapproved by EPA.

2. Is a copy of the approved QA planmaintained at the field site? If not,briefly describe how and whereQA and quality control (QC)requirements and procedures aredocumented at the field site.

T Copies of the QAPP and SSTP areavailable in the trailer. Few operators ofthe trains have used these documents;however, key personnel provided themwith information from these documentsas needed for their tasks.


COMMENTY N

NA


3. Is the design and conduct of theproject as specified in the QAplan? Are there deviations fromthe QA plan? How are anydeviations from the QA plannoted?

T In general, the project is beingimplemented as was specified in theQAPP. However, some procedures havechanged since the revised QAPP wassubmitted to EPA. This wascommunicated to EPA WAM in a letterform on July 8, 1998. The proposeddeviations do not appear to affect thequality of the data being generated.

4. Are written and approved standardoperating procedures (SOPs) usedin the project? If so, list them andnote whether they are available atthe field site. If not, brieflydescribe how and where the projectprocedures are documented.

T No SOPs were observed at the testingsite. EPA test methods are used as theyare and those methods are attached tothe QAPP.

5. For each measured parameter, doany SOPs clearly define the dataquality indicator goals forprecision and accuracy?

T Data quality indicator goals defined inthe EPA methods were used.

6. Do the above data quality indicatorgoals appear to be based ondocumented performance criteriafor the measured parameter or onactual QC data compiled for theparticular measured parameter?

T Data quality indicator goals appear to bebased on EPA methods.

7. Are there established proceduresfor corrective or response actionswhen measurement performancecriteria or the data quality indicatorgoals (e.g. out-of-controlcalibration data) are not met? Ifyes, briefly describe them. Arethey consistent with the QA plan?

T Whenever there are deviations from theQAPP or established procedures in EPAmethods, the EPA Work AssignmentManager was informed by the PES TaskManager.


COMMENTY N

NA


8. For each measured parameter, dothe SOPs specify the frequency ofcalibration, acceptance criteria forthe calibration, and the process forcalibration data reduction andreview?

T PES used the specifications establishedin the EPA methods. All calibration datasheets will be given to RTI at a laterdate. Mr. Dennis Holzschuh, Sr., will beresponsible for QA coordination andreview of the calibration data.

9. Briefly describe how calibrationand other QC data are documented.

T Operational parameters and calibrationdata were recorded on paper data sheets.

10. Does the calibrationdocumentation show thatcalibrations are being performed atthe required frequency and in therequired manner?

11. Are there standard paper orelectronic forms to record QC dataand operational data? Are therecords dated? Is the person whocompleted the record identified? Are paper records written inindelible ink?

T Standard paper data sheets were usedfor all measurements and blank datasheets were supplied to RTI. The datasheets are dated and the operator namewas noted. All data forms were filledout in ink.

12. Are the QC data reviewed byanother qualified person such asthe QA officer or the plantmanager? Who is this individual?

T Mr. Dennis P. Holzschuh, Sr., PES QACoordinator, will review all QA datasheets. He performed internal QA at thefield site.

13. Is the project team adhering to theplanned schedule? If not, explainthe new schedule. Verify that allschedule changes have beenauthorized.

T The schedule is being followed to theextent allowed by unexpectedequipment breakdown and processchanges in the asphalt plant. Allschedule modifications were authorizedby the EPA Work Assignment Manager.

14. Are there written plans to reportchanges in the QAPP during data-gathering activities?

T There are no written plans to reportchanges in the QAPP during sampling.However, the PES Task Managerinformed EPA WAM and other staffverbally concerning schedule modifi-cations and any deviations from theQAPP.


COMMENTY N

NA



B. ORGANIZATION AND RESPONSIBILITIES

1. Identify the following personneland determine whether they havethe listed responsibilities:PES Principal Investigator:

C. Wayne Westbrook • responsible for overall

performance of the project, and• communications with EPA

T Wayne Westbrook was identified asProgram Manager for PES in the QAPP,whereas the SSTP identified JohnChehaske as the Program Manager forPES. In either case, it will not affect thedata quality or scope of the project.

2. PES QA Officer:

Dennis Falgout • assist with and will be

responsible for review andmonitoring of all QA/QCactivities

T Dennis Falgout was not present at theemissions testing.

3. PES Project On-Site Manager:

Frank Phoenix C coordination with PES principal

investigator,• Planning and scheduling the

demonstration project

T Frank Phoenix is the Task Manager andis responsible for all the projectactivities at the emissions testing site.

4. PES On-Site Quality AssuranceCoordinator:

Dennis P. Holzschuh, Sr. C QC activities

T Mr. Holzschuh is responsible for the on-site QA/QC activities and was present.


COMMENTY N

NA


5. Have there been any changes in theproject organization and thepersonnel as outlined in theQAPP?

T In general, there are no significantchanges except staff from AtlanticTechnical Services, Inc., were notpresent. MM5 train samples for theAnalysis of PAHs will be analyzed byQuanterra laboratories instead ofTriangle Labs. This was approved byEPA WAM prior to the field testing.Quanterra labs have high resolutionGC/MS analysis capability for PAHs.

6. Manual Test MethodsOperators:

Mike Maret, VOST at TED

Troy Abernathy, VOST M18 at SED

Dennis Holzschuh, Jr., MM5 at SED

Joe Rubio, MM5& M315 at SED

Brian Purser, M18 at TED

Nick Neilson, M315 at TED

Jessica Swift, MetStation

Jairo Barreda, Shipping & Sample Custodian

T Laura Kinner, GC/MS OperatorJim Peeler, GC/MS OperatorJosh Letomeau, Process Monitor

Each identified operator wasresponsible for sample collection andsample recovery for that method.

7. EPA Work Assignment -ManagerMichael L. Toney

T Mr. Toney was present throughout theentire testing program and coordinatedthe efforts of the emissions testingteams. He authorized all changes intesting schedule and procedures.


COMMENTY N

NA


8. EPA QA Officer:Lara P. AutryC review and approve QAPPC review and approve QA activities

T Ms. Autry was not present at theemissions testing.

9. Does the project maintain a currentsummary of the training andexperience of each individualengaged in the project?

T Training and experience of eachassigned person were maintained at theoffice.

10. Does the project maintaindescriptions of assignmentresponsibilities and requiredproficiency levels?

T Responsibilities are assigned for eachperson and it appears that they are allwell qualified. Only two persons,Jessica Swift and Jairo Barreda, aresummer interns and they have beengiven sufficient training for the workassigned to them.


C. METHOD SPECIFIC - EPA Method 1

1. Is stack diameter properlydetermined?

T The tunnel exhaust is a horizontal 32inch diameter duct which leads from theload-out tunnel to the Smog Hog.

2. Is distance to nearest upstreamdisturbance properly determined?

T

3. Is distance to nearest downstreamdisturbance properly determined?

T

4. Is number of sampling pointsproperly selected?

T For isokinetic testing, a 24-pointtraverse matrix consisting of 12 traversepoints on each of two perpendiculartraverse lines was used.

5. Is points properly marked on pitottube?

T Points are noted on the tape which isattached to the pitot tube and probe.

6. Verification of cyclonic flow isacceptable?

T


COMMENTY N

NA


7. Prior to sampling, all ductdimensions are checked againstpreliminary information to verifymeasurement at locations, locationof test ports, and stack insidedimensions?

T The stack dimensions and locations ofthe traverse points were verified prior tosampling.

8. Are the stack inside dimensions,wall thickness, and sample postdepths measured to the nearest1/16 inch?

T Performed prior to the testing.


The new silo exhaust duct was installed at Silo 2 instead of Silo 3. Since the new SED is 10 inches indiameter, Method 1A was used instead of Method 1. This change was approved by EPA WAM.Some of the work was performed prior to the auditor’s arrival to the site.

The silo exhaust and tunnel exhaust duct dimensions and locations of the traverse points wereverified prior to testing. Prior to testing, the silo exhaust and tunnel exhaust ducts were checked forthe presence of nonparallel flow by recording yaw angle misalignment at each isokinetic samplingpoint. The details are in the PES site-specific Test Plan.

D. METHOD SPECIFIC - EPA Method 2

1. Is pitot tube, lines and manometerassembled correctly?

T A type S pitot tube is connected to aninclined-vertical manometer.

2. Is manometer leveled and zeroedbefore and after each test?

T The box containing the manometers iskept on a bench. Noticed zero beforeand after each run.

3. Is pitot tube checked for chips? Performed leak checks before andafter each test run?

T Leak checks were performed before andafter each test run. Leak rates are notedin data sheets.

4. Is cyclonic flow checked properly? T

5. Is orientation of pitot tube correctduring traverse?

T Noticed the orientation of pitot tube andfound to be acceptable during traverse.

6. Is sampling port adequately sealedaround pitot tube?

T It is sealed either with gloves or a cloth.


COMMENTY N

NA


7. Is process operating at correctconditions?

T Sampling was done whenever theprocess was operating properly and atcorrect conditions (determined prior tothe sampling).

8. Is stable reading taken at eachtraverse point?

T Readings are noted in the data sheets.

9. Is static pressure measured? T

10. Is temperature measured? T The effluent gas temperature is recordedat each traverse point using type-Kthermocouple.

11. Is moisture content determined? Ifso, what method used?

T Moisture content was measured by EPAMethod 315.

12. Is data recorded properly? T Data are recorded in data sheets.

13. Are calculations correct? T Calculations will be performed in theoffice and not at the site.

Additional Questions and Comments:

E. METHOD SPECIFIC - EPA Method 4

1. Is method conducted inconjunction with pollutant testmethods 315 and MM-5?

T Method 315 and MM-5 trains were runsimultaneously both at the silo tunneland silo exhaust.

2. Are impingers properly placed? Impinger contents - 1, 2, 3, 4, 5.

T For Method 315, the first and secondimpingers contained 100 mL of DIwater each, the third impinger is empty,and the fourth and fifth impingerscontained 200 g of silica gel.

3. Sampling time per point? 10 min at the silo tunnel5 min at the silo exhaust

4. Number of points? T 24 points

5. Is probe heater on? Whattemperature? Is it stable?

T Probe is heated at 250 EF. Temperaturereadings are noted in the data sheets.


COMMENTY N

NA


6. Is filter heater on? Whattemperature? Is it stable?

T Filter is heated at 250 EF. Temperaturereadings are noted every 10 min in thedata sheets.

7. Is crushed ice in ice bath aroundimpingers?

T Crushed ice added periodically tomaintain the impingers at 0 EC.

8. Is the exit temperature being keptbelow 68EF?

T Temperatures are periodically recordedin the data sheets (every 5 min).

9. Is pretest leak check conducted? Leakage rate?

T Data sheets will be provided. Leakagerate was within the acceptable limits.

10. Is sampling rate constant? Is itisokinetic sampling?

T Sampling rate is 1.5 ft3/min and foundto be within the specifications.Isokinetic sampling was followed.

11. Is port leak check performed? Leakage rate?

T Data sheets will be provided. Leakagerate was within the acceptable limits.

12. Is electronic balance calibratedwith reading within 0.1 grams ofknown reference standard?

T Calibration was performed in thelaboratory but not at the site. Calibrationdata will be provided.

13. All data are recorded properly? T Noted in data sheets.


Moisture was determined from EPA Method 315 instead of EPA Method 5. Audit questions werecompleted based on EPA Method 315 sampling procedures. Method 4 was performed in conjunctionwith each EPA Method 315 test run.

F. METHOD SPECIFIC - EPA Method 315

Apparatus

1. Is the probe nozzle made of glassor glass lined? What is its design? Is it clean?

T Probe nozzle was made of glass.Cleaned with methylene chloride andmethanol (50:50 mixture). It appears tobe clean.

2. Is the probe liner made ofborosilicate or quartz and is itclean?

T Probe liner is made of borosilicate.Cleaned with methylene chloride andmethanol mixture (50:50). It appears tobe clean.


COMMENTY N

NA


3. Is it type S pitot tube and is itproperly attached?

T S-type pitot tube was used and it isproperly attached.

4. Is the differential pressure gauge astwo inclined manometers?

T The meter box contained two inclinedmanometers.

5. Is the filter holder borosilicateglass; glass frit-support; or siliconerubber gasket?

T The filter holder is a borosilicate glass.

6. Describe number of impingers inthe condenser and its contents?

T The first and second impingerscontained 200 mL of DI water, the thirdimpinger is empty, and the fourthimpinger contained 200 g of silica gel.

7. What type of barometer? Mercuryor Aneroid?

T Aneroid barometer was used. It wascalibrated against mercury manometer.

8. What type of gas densitydetermination equipment used?- sensor type _____- pressure gauge ______Temperature sensor attached toprobe?

T

9. Has filters checked visually forinequalities?

T Glass fiber filter appears to be good.

10. Has filters properly labeled? T Filters after collection were kept in anamber glass bottle and labeled.

11. Has pretest and post-test leakcheck performed properly?

T Leak checks performed properly beforeand after each run and noted in the datasheets.

12. Has impingers properlyassembled?

T Impingers were assembled according tothe Method 315 specifications.

13. Has pitot tube lines checked forplugging or leaks?

T Leak-checked before and after each run.

14. Has meter box leveledperiodically?

T Meter box was kept on a horizontalsurface.

15. Have manometers zeroed? T


COMMENTY N

NA


16. Has acetone-insoluble head-stablesilicone grease added to all groundglass joints?

T No silicone grease was used to theground glass joints. They are used onlywhen it leaks.

17. Is probe heat uniform along lengthof probe?

T Probe is kept at 250 EF and thetemperatures are noted in the data sheetsevery 5 min.

18. Has effective seal made aroundprobe when in-stack?

T Sealed with hand gloves or a cloth andfound to be acceptable.

19. Nozzle & pitot tube parallel tostack wall at all times.

T

20. Has filter changed during run? Any particulate lost?

T No filter changed during the run. Noparticulate was lost.

21. Have data sheets complete and dataproperly recorded?

T Data sheets are filled out properly witha legible ink pen.


1. Due to problems in the silo tunnel and silo exhaust, the samples collected on the first day(7/23/98) were voided after approval from EPA WAM. After 2 hr sampling, it was noted thatthe silo exhaust damper was closed. Therefore, the damper was operated and continued thesampling for another 2 hr. The first day run will be repeated on 7/27/98.

2. Observed the Method 315 sampling run at the silo 2 exhaust at the roof on 7/25/98. Flow rate0.75 ft3/min. Sampling was done only for 2 hr due to expected high concentrations. Method 315and MM5 trains were operated at two different times. Due to smaller stack diameter, traversereadings were taken from one place and probe was kept in another hole and readings were notedfor every 5 min. Stack is at negative pressure. Stack hole is covered with gloves.

3. 4-hr sampling was performed from the silo 2 tunnel while two trucks were going through thetunnel to get the background emissions.

LABORATORY INFORMATION NECESSARY FOR EPA METHOD 315

1. Was the calibration informationdocumented in log books?

T

2. Have the previous calibrations metthe acceptable tolerances?


COMMENTY N

NA


3. Has the Type-S pitot tube beenverified using standards outside ofthe laboratory?

4. Describe the pitot tube calibrationapparatus.


Calibrations were performed in the laboratory prior to the testing on the site. Calibration sheets willbe supplied to the auditors after the field test.

Calibration and General

1. Have the dry gas meters beencalibrated against a standard? (Ifso, which standard?)

T

2. Have the stack and traintemperature sensors beencalibrated against a referencethermometer?

T

3. Has the nozzle been calibrated tonearest 0.025 mm (0.001 in), andhas the nozzle size been properlyselected?

T

4. Is the train correctly setup up andleak-checked to less than 4% or0.00057 m3/min (0.02 ft3/ min),whichever is less?

T Leak checks were performed before andafter the sample collection and found tobe acceptable.

5. Are the train components clean andfree of breaks, cracks and leaks,and is the probe liner clean andleak free at 380 mm (15 in) Hg?

T

6. Are the pitot tube lines free ofplugs or leaks?

T

7. Is the probe heating systemoperating correctly?

T


COMMENTY N

NA


8. Are the pitot tube and temperaturesensors properly attached to probe?

T

9. Are the nozzle and pitot tubelocated parallel to stack wall? Observe handling and positioningof the probe.

T

10. Is the filter holder temperaturemaintained at 120 ± 14EC (248 ±25EF) throughout the tests?

T

11. Is the sample gas temperature atthe last impinger maintained at <20EC (68EF) throughout the tests?

T

12. Is isokinetic sampling maintainedwithin ± 10% and checked everyfive minutes?

T

13. Is the sample site selection anappropriate distance downstreamfrom any flow disturbance?

T

14. Are filters free of irregularities,properly installed, and properlylabeled?

T

15. Do the equipment operators haveaccess to test protocols andmethods, are data sheets available,and is equipment in good repair.

T Equipment operators have copies of theprotocols and methods. Data sheets areavailable to each train operator.

16. Are all data recorded andcalculations checked? Is at leastone decimal point beyond that ofacquired data retained?

T Calculations will be checked in theoffice.

17. Are all impingers cooled in an icebath at all times?

T Crushed ice was added periodically tomaintain the temperature.

18. Is full stack traversing beingconducted?

T

19. Do recovery methods preventcontamination?

T Sample recovery was performed in thetrailer, which is relatively clean.


COMMENTY N

NA


20. Are graduated cylinders within 2ml (or less) subdivisions used?

T Graduated cylinders within 2 mLsubdivisions were not used. However,standard impingers were used and aknown volume of solutions was added.

21. Are all openings capped duringtrain disassembly?

T Capped with Teflon tape.

22. Is the clean-up area clean andprotected from the wind?

T It was performed in the trailer. The areais relatively clean.

23. Are any particulates spilled? T The operators are very careful intransferring the filter paper to the glassjar.

24. Do any visible particles remaininside the probe?

T None was noticed inside the probe.

25. How is the pitot tube cleaned? T Cleaned with methylene chloride andmethanol (50:50) mixture.

26. Are sample containers labeled andsealed tight, and is the liquid levelmarked?

T Electronically printed labels wereaffixed to the containers. Liquid level ismarked with a pen and sealed withTeflon tape.

27. What is the history of the S-typepitot being calibrated against a stdpitot?

The calibration was performed in thelaboratory and the calibration data willbe supplied to the auditors after the fieldtest.

28. How was the inside diameter of thestack measured?

It was measured prior to the testing andnoted to be 32 inches.

29. How is the probe marked to ensureproper sampling locations?

T Probe is wrapped with tape at eachtraverse point and marked with a pen.

30. How long does the probe remain ateach sample point? Is the readingstable?

T The probe remained at 10 min at eachsample point. Readings are noted in thedata sheets.

31. How is alignment ensured whilesampling? Guidance manualsspecify visual observations and notthe highest delta p.

T Visual observations indicate thatalignment appears to be QC.


COMMENTY N

NA


32. Were the unused ports pluggedwhile sampling occurred?

T Unused ports were sealed with electrictape.

33. Was the equipment damagedduring the test? Explain.

T A few of the glass nozzles were brokenduring recovery operation. They werereplaced with new ones for the next run.

34. How is static pressure measured,and what is it? ...sampling?

Static pressure was measured with aninclined manometer.

35. How is steady state processindicated during sampling?....would be the impact?

T It will not be a steady-state process.Impact will be minimal due to anintegrated sample.

36. Were they any changes in thefacility emissions during sampling?

T Emissions may vary at the facility dueto the nature of the operation.

37. If changes in emissions did occurduring sampling, what was theimpact?

Impact will be minimal due to anintegrated sample collection method.


G. METHOD SPECIFIC - EPA Method 18

1. Have you performed presurveysampling using Method 18procedures?

T Presurvey sampling was not performedusing Method 18 procedures. However,grab samples were collected andanalyzed using canisters. Method SW846 M0030 VOST data were collectedat the hot mix dryer stack.

2. What are the approximateconcentrations of the targetedcompounds?

3. What is the sampling rate? Is itconstant or proportional?

T Sampling rate was 1 L/min and sampledfor 4 hr at silo tunnel, whereas 2 hrsampling was done at silo exhaust.

4. What is the sample time? T 4 hr at the silo tunnel, 2 hr at the siloexhaust.


COMMENTY N

NA


5. Probe heat or not. If so, whattemperature?

T Not heated. Teflon lines were used.Ambient temperature.


Apparatus

6. What type of probe is used—stainless steel, glass or Teflon?

T Teflon lines were used.

7. Is it heated and any filter used? T No filter was used and Teflon lines arenot heated.

8. Silica gel tube or extra adsorptiontube used prior to adsorption tubewhen moisture content is > 3percent.

T None was used at the silo 2 tunnelemissions sampling. However, twoimpingers at ice path temperature wereused at the silo exhaust.

9. Is the leakless sample pumpcalibrated within limiting (sonic)orifice or flow meter?

KNS Teflon pump was used to pull thesample gas. The pump was calibrated inthe laboratory with a bubble flow meter.

10. Is rotameter used to detect changesin flow?

T

11. Are the two sampling trains run inparallel?

T The gas is pulled through a T-connection into spiked and unspikedsampling tubes.

12. Are the pre-spiked adsorbent tubesused in one train and non-spikedadsorbent tubes used in the secondtrain?

T Two different probes but pulled througha common pump.

13. What are the compounds spikedand at what concentrations?

T 200 µg each of benzene, toluene,xylene, hexane, and cumene was spiked.Liquid mixture was spiked onto thecartridge prior to sampling.

14. Is accurately measured sample timewith a stop watch?

T


COMMENTY N

NA


15. Is the sampling pump and flowmeter recently calibrated withbubble meter?

T

16. Whether extreme care is taken toensure that no sample is lost in theprobe or sample line prior to theadsorption tube.

T

17. Is pretest leak check acceptable?(No flow indicated or meter.)

T

18. Total sample time, sample flowrate, barometric pressure, andambient temperature are recorded?

T Recorded in data sheets.

19. Total sample volumecommensurate with expectedconcentration and recommendedsample loading factors?

T 4 hr sampling was performed at the silotunnel. 2 hr sampling at the silo exhaust.

20. Is post-test leak check and volumerate meter check acceptable? (Noflow indicated on meter, post-testcalculated flow rate is within 5percent of pretest flow rate?)

T The gas sample is pulled into spiked andunspiked sampling tubes.

21. Is desorption efficiency determinedfor adsorbent to be used for fieldsampling? If so, what are thedesorption efficiencies?

T The analyst reported that the desorptionefficiencies were determined in the labprior to the testing and they are 95percent for each compound.

22. Is collection efficiency determinedfor adsorption tubes used for actualfield sampling?

The analyst reported that the collectionefficiencies were determined in the lab.The acceptable range is 70-130 percent.However, 85 percent was found.

23. Are the adsorbent tubes storedproperly after sampling?

T They were stored in an ice cooler aftersampling.


COMMENTY N

NA



AnaSorb 747 (1,000 mg front + 1,000 mg backtrap) was used. Carbon disulfide + dimethylformanide will be used for desorption. No glass wool was used in the Teflon lines. Sample gas ispulled through a vacuum pump and maintained a flow through a critical orifice. There may be asmall pulsation and the impact is minimal. Method 18 was performed for two losses at 1 L/min at thesilo exhaust. Impingers were used prior to the sampling tubes at the silo exhaust to condense thewater vapor.

H. METHOD SPECIFIC - SW-846 Method 0030 (VOST)

1. Will the total run time be aminimum of 2 hours?

T The total run time is 20 min. Four runswere performed on each day.

2. Is the maximum load time 20 min.For each pair of cartridges? At1L/min sampling rate?

T Generally, 1 L/min sampling rate wasused at silo 2 tunnel emissions.However, 0.25 L/min for 10 min wasused at the silo exhaust emissions.

3. Is a quartz liner wed inside thestainless steel probe?

T Glass wool was inserted in the front ofthe quartz liner to remove anyparticulates.

4. Is the probe maintained at atemperature greater than 150EC?

T Probe temperature was maintained at255 EF.

5. Do the sorbent cartridges appear tobe sealed tightly in their carrycases?

T The sealed cartridges were kept incarrying cases.

6. Have the sorbent cartridges beenplaced correctly in the sample train(1st cartridge Tenax; 2nd cartridgeTenax/charcoal)?

T

7. Do all vacuum gauges, pumpconnections, calibrated rotameters,and dry gas meters appear in goodworking order with leak-tightseals?

T


COMMENTY N

NA


8. Is there an available thermal couplereading between the first condenserand first Tenax cartridge? CHECKTEMPERATURE. MUST BELESS THAN 20EC.

T Temperature was measured and it is lessthan 20 EC.

9. Is the sample line between theprobe and the first condenser lessthan 5 ft.?— Is it Teflon?— Is it heat traced to 150EC?— Are compression fittings used?

T Teflon line was used. It is not heattraced.

10. Does each individual cartridgehave its own unique ID numberand flow direction arrow markedon it?

T

11. Were the sorbent cartridgesreceived in the field on cold packs(~4EC)?

T

12. Is the cooling water to thecondenser being maintained at ornear 0EC?

T Higher than 0 EC, but lower than 20 EC.Ice water circulated.

13. Is the train leak-checked at thebeginning of a run to the followingspecifications? (1) pull a vacuumto 250 mm Hg (10 in Hg), (2) leakrate less than 2.5 mm Hg after 1minute?

T

14. Is a charcoal canister used to filterthe ambient air used to returnsystem to ambient pressure afterthe leak check?

T

15. Was the sample train leak-checkedafter 20L of gas was sampled andwas the highest pressure dropencountered recorded so that hisvalue can be used during thesecond leak test?

T Leak test was performed before andafter each run. The sampling rate andvolume are different at the silo exhaustemissions testing. It was approved bythe EPA WAM.


COMMENTY N

NA


16. Are the field blanks exposed forthe same amount of time as it takesto change sorbent cartridges in theVOST?

T Field blank was performed on 7/25/98.

17. Was there a field blank exposed foreach run?

T Field blank was performed on 7/25/98.


7/25/98. Due to higher concentrations at the silo exhaust, it was authorized to run VOST at 0.25L/min for 10 min. Four runs were done each day. Stack is at negative pressure. No moisturecondensation was observed in the impinger. The flow rate of 0.25 L/min is at low setting in theVOST console box. Rotameter setting is 0.2, which is the bottom of the scale.

I. METHOD SPECIFIC - SW-846-0010 (Modified Method 5)

1. Is the probe nozzle made of 316stainless steel or glass with a sharp,tapered (30E angle) leading edge?

T The probe nozzle is made of glass.

2. Is the nozzle of buttonhook orelbow design and seamless?

T The nozzle is buttonhook.

3. Is the nozzle calibrated? T It is calibrated in the laboratory.

4. Is the probe glass lined and heatedto 120 ± 14EC at the exit endduring sampling?

T

5. Is the type s pitot tube on a planeeven with or above the nozzle entryplane during sampling?

T

6. Does the pitot tube have a knowncoefficient? What is it?

T The coefficient is 0.84.

7. Does the train have twomanometers: one for velocity-head()P) readings and one for orificedifferential pressure readings()H)?

T Two inclined manometers were used.

8. Is the filter holder made of glass,and the gasket of Teflon?

T Glass


COMMENTY N

NA


9. Is the XAD-2 module cooled to 17± 3EC?

T Ice water bath was maintained.

10. Is there a thermocouple in the glasswell of the sorbent module tomonitor temperature of the trap?

T The temperature is 42 EF.


Train Components

1. Is the sorbent module followed bya condensate knockout trap?

T

2. Is stack gas moisture beingmonitored with four 500 mLimpingers following the knockouttrap? List the impinger solutions.

First impinger is empty. Second andthird impingers contained 200 mL ofHPLC-DI water. Fourth impinger isempty. Fifth impinger 200 g of silicagel.

3. Is the metering system consistingof vacuum gauge, leak-free pump,thermometer, and dry gas meter,keeping the sampling rate of theMM5 train within 10% ofisokinekticity?

T

4. Is the metering system determiningthe sample volume within 2% ofthe true volume?

T

5. Is the barometric pressure beingrecorded and corrected forelevation differences between thelocation of the barometric and thesampling point at a rate of -2.5 mmHg per 30-m elevation increase (or+2.5 mm Hg per 30-m elevationdecrease)?

T Barometric pressure is noted from theMet Station and at the truck.


COMMENTY N

NA


6. Is the stack gas temperature sensorattached to the pitot tube, such thatthe tip of the sensor extendsbeyond the leading edge of theprobe sheath and does not interferewith the pitot tube opening?

T

7. Are calibration and preparationdata being recorded in apermanently bound laboratorynotebook (i.e., filter and silica gettare weights, clean XAD-2,QA/QC check results, dry-gasmeter, and thermocouplecalibrations)?

T Noted in the data sheets but not in thebound notebook.

8. Is the train grease-free upstream ofthe module?

T


Preparation of the Train

1. Are the probe liner brushesprovided at least as long as theprobe?

T

2. Are the probe brushes made ofnylon bristles and stainless steelwire handles with extensions?

T Probe brushes were made of nylonbristles.

3. Are the wash bottles made ofTeflon or glass?

T Teflon bottles were used.

4. Are the sample storage bottlesmade of amber glass and screw-type Teflon lined caps?

T Amber glass bottles with Teflon-linedcaps were used.

5. Are the filters made of glass orquartz fiber?

T Glass fiber filters were used.

6. Has the XAD-2 resin been cleanedprior to use and QC to a blank lessthan 4 mg/kg?

T Quanterra labs provided the XAD-2cartridges.


COMMENTY N

NA


7. Is distilled organic-free water(Type II) provided for the impingersolutions?

T HPLC grade DI water was used.

8. Are the clean up solvents pesticidegrade?

T Methylene chloride + methanol (50:50)pesticide grade solvents were used.

9. Are the filters desiccated at 20 ±5.6EC and ambient pressure for atleast 24 hours, weighed at intervalsof at least 6 hours to a constantweight (± 0.5 mg), and recorded at0.1 mg?

T Weight of particulates is not measuredwith MM5 train.

10. Have the number of minimumsampling points been determinedfor each sampling location (inlet,outlet, and stack): What are thenumbers?

T 24 points.

11. Have the pitot lines been leakchecked?

T Leak checked before and after the runand found to be leak free within thespecifications.

12. Has the stack gas moisture contentbeen determined? What is it?

T Stack gas moisture content at the silotunnel is -1-2 percent. The moisture ishigher at silo exhaust.

13. Were there any nozzle changesduring sampling?

T No nozzle changes during sampling.Nozzle diameter is 0.252 inches.

14. Is the sample volume as outlined inthe sampling plan being pulled? What is it?

T 0.75 ft3/min for 4 hr at the silo tunnel.0.75 ft3/min for 2 hr at the silo exhaust.

15. Are all openings to the train keptcovered with Teflon film prior toassembly?

T

16. Are the XAD-2 modules assembledoff-site?

T XAD-2 modules were supplied byQuanterra labs. Assembled in the rack.

17. Is the filter placed in the filterholder with tweezers or whilewearing surgical gloves?

T


COMMENTY N

NA


18. Is a pre-test leak check of the trainadministered?

T Noted in the data sheets.

19. If a train component is changedduring sampling, is the train leak-checked prior to the change? Is theleakage rate less than 4% of theaverage sampling rate?

T No components were changed duringsampling.

20. Is a post-test leak checkperformed? Is the leakage rate lessthan 4% of the sampling rate?

T Leak rate is 0.02 inch.


Running the Train

1. Is the sampling rate maintainedwithin 10% of true isokinetic?

T

2. Is the filter maintained at 120 ±14EC?

T

3. Is the sorbent module maintainedat 17 ± 3EC?

T Ice water temperature was maintained.The module is covered with aluminumfoil.

4. Are the dry gas meter readingsrecorded on field data sheets?



Sample Recovery

1. Is the filter removed from the filterholder and placed in an identifiedPetri dish container with tweezers?

T The filter was kept in an amber glassjar.

2. Is the filter folded with the filtercake inside?

T The filter is folded and kept in an amberglass jar.


COMMENTY N

NA


3. Is any particular matter adhering tothe filter-holder gasket transferredto the Petri dish?

T The particulate matter adhering to thefilter-holder gasket was transferred intothe glass jar.

4. Is the Petri dish sealed and labeledcontainer No. 1?

T The amber glass jar is sealed andlabeled.

5. Is the particulate matter from theprobe nozzle, probe fitting, probeliner, and front half of the filterholder recovered quantitatively bywashing with (1:1 v/v)methanol/methylene chloride into aglass container? Labeled containerNo. 2?

T Collected into amber glass containerand labeled. The liquid level is markedon the container with a pen.

6. Is a solvent blank retained andanalyzed?

T

7. Is the sorbent module labeledcontainer No. 3, capped, andpackaged cold for shipment?

T The sorbent module weighed, labeled,and kept in blue ice.

8 Is the condensate from the backhalf of the filter module and theknockout trap measured andtransferred to container No. 4?

T Collected into amber glass jar andlabeled. The liquid level is marked onthe jar with a pen.

9. Are the sample train componentsbetween the filter and the first wetimpinger rinsed into container No.5?

T

10. Is the silica gel transferred tocontainer No. 6?

T Weighed for the moisture content; silicagel changed for each train.

11. Is the volume of the impingerwater in the first three impingersmeasured to ± 1 ml and any coloror film noted? Is the impingercatch discarded?

T No color or film was noticed. Theimpinger catch is saved in the glass jars.

12. Are all containers sealed and allshipped refrigerated except thefilter?

T


COMMENTY N

NA



J. SAMPLING GENERAL

1. Is the duration of samplingsufficient to detect all importantpollutant(s) generated by theprocess under investigation?

T Sampling for 4 hr at the silo 2 tunneland 2 hr at the silo exhaust is quitesufficient.

2. Are there specified limits forwhich sampling will be stopped? Describe.

T Sampling will be stopped only whenloading or process problemsexperienced at the site.

3. Were duplicate or replicatesamples taken for each samplinglocation?

T One run was performed on each day.

4. Was the sampling performed inaccordance with the approvedQAPP?

T Generally followed the approved QAPP.However, few changes such as samplingrate at the silo exhaust testing weremade at the site after EPA WAMauthorization.

5. Do sample tracking numbersindicate when, how, and where thesamples were collected?

T Labels were attached to the glass jars.

6. Are records available of whocollected the sample?


7. Are field log books and laboratorylog notebooks recorded in ink?

T Data sheets were used. Recorded legiblywith ink.

8. Are audit histories recorded in thefield log notebooks?— thermocouple— solid state temperature— wood scale performance— other

T Noted in the data sheets in the office.



COMMENTY N

NA


K. QUALITY ASSURANCE/QUALITY CONTROL/GENERAL

1. Have there been any changes in theproject organization and thepersonnel as outlined in theQAPP?

T

2. Are there written plans to reportchanges in the QAPP during data-gathering activities?

T

3. Were field biased blanks and/ortrip blanks used?

T One field blank was collected.

4. Calibration information and datasheets available for:— Probe/nozzle— Type-S pitot— Meter Box including: dry gas meter and thermocouple— Barometer

T Data sheets will be given.



COMMENTY N

NA


L. SAMPLE CUSTODY AND INTEGRITY

1. Is there an SOP or other source ofdocumentation which describes theorganization’s sample custodyprocedures?

T There is no SOP at the site. However,sample custody data sheets andprocedures are available.

2. Are records available of when,how, and where the sample wascollected?

T Data sheets are available. Information isnoted in the data sheets and labeled onthe bottles.

3. Are records available of whocollected the sample?

T Recorded on the data sheets.

4. Who is the field custodian for thesamples? (Name)

T Jairo Barreda, a summer intern, isresponsible for the shipping and samplecustody.

5. Are samples individually identifiedby number or code so that they canbe traced?

T Labels are electronically made, printedfor each run with identification, andaffixed to the glass jars.

6. Were deviations (if any) from allSOPs or protocols properlydocumented?

T Recorded by the Task Manager in asmall notebook. It would have beenbetter if the train operators noted on thedata sheets also.

7. Sample liquid levels marked oneach container such that the markcan be seen on the container or onthe label itself.

T Marked with a pen on the container.

8. Sample container lids sealed onoutside with Teflon tape andsample label covered with solventresistant tape?

T Solvent-resistant tape is not used tocover the sample label.

9. Are integrity seals affixed over thecaps of each container?

T Labels are affixed on the side of theglass jar.

10. Are seals signed and numbered byrecovery person?

T Seals are not signed. Numbered on thelabels.



COMMENTY N

NA


M. MISCELLANEOUS MEASUREMENTS - Wind Speed, Wind Direction, AmbientTemperature and Humidity

1. Is the meteorological stationoperated according to themanufacturer’s specification?

T

2. Is the manufacturer’sspecifications manual available atthe site?

T

3. Are the readings recordedcontinuously through out the testperiod?

T Recorded on a data-logger.

4. Was the wind speed measured andcalibrated?

T Calibrated at the office but not at thesite.

5. How the wind direction measuredand calibrated?

T Calibrated at the office but not at thesite.

6. How the temperature is measuredand what is the range?

T Thermometer -50 to 50 EC

7. Is the ambient humidity measuredonce per test run by wet bulb usinga sling psychrometer and properlyrecorded?

T Measured once in the beginning andonce at the end of the day.


The MET Station was kept at the top of the load-out tunnel near the entrance.

N. Estimation of Particulate Deposition on the Ceiling of the Load-Out TunnelDownstream and Silo No. 5

1. Is one test plate for each section isinstalled properly in the beginningof the test program?

T

2. Are the test plates and ceilingsections cleaned with acetoneproperly in the beginning and atthe end of the test program?

T


COMMENTY N

NA


3. Have they collected three acetonerinse samples and analyzedgravimetrically?

Analysis will be performed in the office.


Measured the distance from Silo No. 5 to the end of the tunnel which is approximately 57 feet.Divided the distance into three equal areas. Three test plates and ceiling sections were cleaned withacetone. Each test plate is screwed to the top of the roof at 19 feet distance from one another on7/2/98 morning. They were removed on 7/26/98 and rinsed with acetone and will be analyzedgravimetrically.

O. Estimation of Particulate Deposition on the Inside Walls of Loud-Out-TunnelExhaust Plenum

1. Are six pieces of box pipe installedproperly inside the exhaust plenumunder Silo No. 3 in the beginningof the test program?

T Charged under Silo No. 2 on 7/23/98and removed on 7/26/98 morning.

2. Are the box pipes and installationsections cleaned with acetoneproperly in the beginning and atthe end of the test program?

T Installation sections were not cleaned atthe end of the test program.

3. Have they collected six acetonerinse samples and analyzedgravimetrically?

T Acetone rinse samples will be analyzedlater in the laboratory.


Installed at the downstream of the exhaust plenum.

P. Estimation of Particulate Deposition on the Inside Walls of the TemporarySilo Exhaust and Wind Tunnel Exhaust

1. Have they collected one acetonerinse sample from inside walls ofthe temporary silo exhaust and foursamples from the wind tunnelexhaust?

T One was kept at the silo exhaust andthree samples were taken from the windtunnel exhaust.


COMMENTY N

NA


2. Have they cleaned the areas andinstalled the test plates properly?

T


Old silo exhaust pipe was cleaned and taken for metals analysis.

Q. Direct Interface Portable GC/MS

1. Is heated sample line used to drawthe sample into the GC/MS?

T Teflon lines were used at ambienttemperature.

2. How the instrument wascalibrated?

T Calibrated with 36 compounds in thelaboratory. Checked with 9 componentmixture at the field site daily.

3. What is the flow rate and samplevolume collected?

T Sample is drawn at 2 L/min.Subsequently 150 cm3/min is divertedinto the mass spectrometer. A 20-30 µLsample was used to inject into the GC.

4. How is the sample concentrated? T No concentration step.

5. Is the thermal desorption techniqueused and at what temperature itwas desorbed?

T No concentration step. Therefore,thermal desorption is not required.

6. Are there blank runs performed? T Nitrogen blank daily was performed.

7. How is the moisture interferencethat was eliminated or minimizedduring the analysis?

Up to 8 percent moisture levels, noconditioning is required. Above that daythe sample used an impinger beforeintroducing it into GC/MS.


1. GC/MS runs were done at silo tunnel, silo exhaust, ambient air, end of the tunnel, and dryerexhaust.

2. Lot of moisture was condensed at the silo exhaust and dryer exhaust. Impinger was used tocollect the water. Linearity checked in the laboratory, 6 injections ±20 percent RSD.


COMMENTY N

NA



Column used: SPB-1 3-inch wound capillary column sample drawn through 0.3 µm glass fiber filter.Detection limits vary from compound to compound.

This

is

a

Blank

Page

TECHNICAL REPORT DATA(Please read Instructions on reverse before completing)

1. REPORT NO.

EPA-454/R-00-0262. 3. RECIPIENT'S ACCESSION NO.

4. TITLE AND SUBTITLE

HOT MIX ASPHALT PLANTS TECHNICAL SYSTEMS AUDIT of TESTING at PLANT C ASPHALT PLANT C LOS ANGELES, CALIFORNIA

5. REPORT DATE

May 2000

6. PERFORMING ORGANIZATION CODE

7. AUTHOR(S)

Lara Autry (EPA) R.K.M. Jayanty (RTI) Robert S. Wright (RTI)

8. PERFORMING ORGANIZATION REPORT NO.

9. PERFORMING ORGANIZATION NAME AND ADDRESS

Research Triangle Institute Center for Environmental Measurements and Quality Assurance 3040 Cornwallis Road, Post Office Box 12194 Research Triangle Park, NC 27709-2194

10. PROGRAM ELEMENT NO.

11. CONTRACT/GRANT NO.

68-D4-0091

12. SPONSORING AGENCY NAME AND ADDRESS

Office of Air Quality Planning and Standards Office of Air and Radiation U.S. Environmental Protection Agency Research Triangle Park, NC 27711

13. TYPE OF REPORT AND PERIOD COVERED

Final

14. SPONSORING AGENCY CODE

EPA/200/04

15. SUPPLEMENTARY NOTES

16. ABSTRACT

The United States Environmental Protection Agency (EPA) Emission Factors and Inventory Group (EFIG) isinvestigating the Hot Mix Asphalt industry to identify and quantify criteria and hazardous air pollutants (HAP’s) emitted fromtransport truck loading and silo filling. EFIG requested that EPA’s Emission Measurement Center (EMC) conduct the requiredtesting. Under separate EPA contracts, Midwest Research Institute (MRI) and Pacific Environmental Services (PES)performed this emissions testing. An independent technical systems audit was performed as part of the quality assuranceevaluation of this testing effort. The EMC issued a work assignment to Research Triangle Institute (RTI) to perform thistechnical system audit. The primary objective of the testing program was to characterize uncontrolled emissions of the criteriapollutants particulate matter (PM) and total hydrocarbons (THC) and emissions of volatile and semi-volatile organic HAP’sincluding polycyclic organic matter, phenol, benzene, toluene, xylene, ethylbenzene, 2-butanone, cumene, formaldahyde,hexane, isooctane and others.

This report includes a summary of the independent audit results and detailed Technical Systems Audit checklists ofthe evaluations RTI performed of the testing by MRI and PES.

17. KEY WORDS AND DOCUMENT ANALYSIS

a. DESCRIPTORS b. IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group

Hot Mix AsphaltSilo FillingTruck LoadingParticulate MatterVolatile Organic CompoundsTotal HydrocarbonsHazardous Air PollutantsPolycyclic Organic Matter

Air Pollution control

18. DISTRIBUTION STATEMENT

Release Unlimited19. SECURITY CLASS (Report)

Unclassified21. NO. OF PAGES

14220. SECURITY CLASS (Page)

Unclassified22. PRICE

EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE

This

is

a

Blank

Page

Date post:	01-Jun-2021
Category:	Documents
Upload:	others
View:	8 times
Download:	3 times

HOT MIX ASPHALT PLANTS - EPA...Hot Mix Asphalt Plants Technical Systems Audit of Testing at Asphalt...

Documents