American Association for Aerosol Research - Abstract Submission

AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA

Abstract View


Assessment of PM Measurements Used in the US-EPA Residential Wood Heating Appliance Emission Test Method

GEORGE ALLEN, Lisa Rector, NESCAUM

     Abstract Number: 467
     Working Group: Instrumentation and Methods

Abstract
The US EPA issued revisions to the New Source Performance Standards (NSPS) for residential wood heaters in February, 2015 that tightened PM emission limits for new wood heaters. Changes to regulatory test methods were part of this NSPS rule, and included the option of using Teflon coated glass fiber filter media in addition to the glass fiber media that has been used for several decades. The addition of Teflon media was in response to concerns regarding possible acid gas artifacts when glass fiber filters are used for measuring PM emissions for residential wood combustion, which could bias test results and degrade reproducibility of tests. While EPA test method 5 specifies use of inert filter media when sampling stack PM containing acidic gases, methods 5G and 28 for wood heaters specified glass fiber filters (a basic media) be used. The implicit assumption that woodsmoke does not have acidic gases is not always correct, at least for eastern wood like red oak which can have more than 100 ppm sulfur content and thus SO2 in the stack gas. Twenty-six sets of gravimetric dilution tunnel PM samples from a 1988 NSPS-compliant wood stove burning red oak were collected using glass fiber, Teflon coated glass fiber, and Teflon membrane filters under a range of burn conditions. PM sample concentrations ranged from 3 to 200 milligrams/m$^3; sample duration was usually between 15 and 90 minutes. Teflon membrane and Teflon coated samples agreed well (slope=1.06) and were well correlated (R2=0.97). Glass fiber samples were inconsistently higher than Teflon coated samples (10 > 30%, 2 > 200% higher), with slope=1.19, R2=0.89. These results are consistent with a positive acid gas artifact on glass fiber filter media.