2.4. Field Gas Chromatographs (GCs)

Emerging Field GCs for NGEM applications range from fence-mounted low-cost units (< $25,000 USD) with single source detection goals, to self-contained mid-cost units (< $100,000 USD) possessing improved detection limit and speciation capabilities, to higher performance systems housed in traditional walk-in air monitoring shelters. The first year of this project successfully deployed a mid-cost field GC (MiTAP^® P310, Tricorntech Corp. Taipei City, Taiwan; Figure 2c) and an early configuration of the auto-GC system, customized for Rubbertown-specific compounds (Airmo VOC C3-C6, AirmoVOC C6-C12, Chromatotec America Inc., Houston, TX, USA; shelter-integration by Consolidated Analytical Systems, Cleves, OH, USA; Figure 2a), deployed by LMAPCD. The deployment of a prototype low-cost benzene GC was also briefly attempted but produced no usable data.

The small form factor MiTAP GC was housed in a weather-proof enclosure (Figure 2c), containing carrier and calibration gas, an automatic calibration check system, and an ACGS trigger system allowing ~one-minute duration samples to be acquired in a 1.4-liter Silonite^® canister via GC trigger. The MiTAP sampled air for 15 minutes of each hour, producing 22 hourly values per day, with the calibration check taking two hours per day. The MiTAP GC was configured to quantify 12 compounds, (selected 1,3-butadiene results discussed here), with five other compounds tracked as near coeluting interferents. The on-board calibration cylinders of the MiTAP (five in total) contained between 4.6 ppbv and 8.6 ppbv 1,3-butadiene, as determined by lab testing. A conservative manufacturer-set MDL was 0.1 ppbv for this compound (the reporting limit). The MiTAP GC and calibration cylinder concentrations were tested at the EPA VOC laboratory and were found to have a low bias; however, due to the extended range of data presented here (well above calibration range with uncertainties in linearity), no additional bias correction factor has been applied.

In its initial deployment, the LMAPCD auto-GC system was set up to produce two data points per hour with air sampled for 22.5 minutes of each 30-minute period. The GC used for 1,3-butadiene measurements employed permeation tube checks with n-butane, n-hexane, and benzene sequentially twice per day to measure retention times. Accuracy comparisons to EPA-prepared standards in the 5 to 10 ppbv range were executed on several occasions. The LMAPCD GCs have the potential to automatically quantify dozens of VOC and HAP compounds relevant to Rubbertown, but the initial deployment of the system found significant issues with stable operation caused by factors including, but not limited to, humidity-related retention time shifting and difficulties with coeluting compounds exacerbated by proximity to a fuel storage terminal and waste water treatment facility.

The chromatography of 1,3-butadiene was sufficiently separated from interfering species, so it could be positively identified and integrated with confidence. The 1,3-butadiene MDL for this analysis was estimated at 0.1 ppbv. Due to the prototype nature of both GC systems, data availability for the first year of deployment required significant post processing and expert analysis.