Real Time Detection Systems II

Podium Se​ssion 123

Wednesday, June 4, 2014, 10:00 AM - 12:00 PM 

SR-123-01 Field Measurement of Common Fumigants via Portable GC with PID & FUV Detectors 

J. Maclachlan, PID Analyzers, LLC, Sandwich, Massachusetts, MA 

Objective: More than two dozen different chemicals have been used as fumigants for grain including organics like methyl bromide, inorganic chemicals like sulfuryl fluoride ,and even chemical agents such as cyanogen chloride. Deaths from fumigants have occurred in rail cars, ships, grain elevators, green houses, pest control … A field method for detecting common fumigants is essential since many workers encounter these dangerous chemicals which do not have OSHA field detection methods. 

Methods: Utilizing a portable gas chromatograph Model GC312 equipped with a battery and weighing in at 26 pounds, this field tool can detect part per billion (ppb) levels of fumigants with a photoionization detector (PID). A 30M x 0.32 mm capillary with 5 or 10 micron methyl silicone film is used for separating gases and volatile compounds. The far UV absorbance (FUV) detector detects the low or sub ppm levels of a variety of hydrocarbons & low molecular weight (MW) gases such as hydrogen cyanide, sulfur dioxide, and sulfuryl fluoride that absorb at 126 nano meters (nm). 

Results: Some of the common fumigants used today are methyl bromide, ethylene oxide, phosphine, and ethylene dibromide. All of these compounds can be detected at ppb levels by GC/PID with a thick film capillary column. The PID has a detection limit of < 5 ppb for PH3, and can detect part per trillion (ppt) levels of dibromochloro propane (PEL = 1ppb) by PID if the internal concentrator on the portable GC is used. The FUV can detect hydrogen cyanide, ethylene dichloride, dichloropropane, sulfur dioxide and sulfuryl fluoride at low or sub ppm levels. The detection limit for SO2 by FUV is 0.5 ppm. 

Conclusions: A portable GC with a thick film cap column equipped with PID and Far UV detectors can detect the most common fumigants in use today at levels of 0.1 of the PEL. The portable GC is easily carried to the field and can be operational in about 2030 minutes. It can detect the ten most common fumigants described above plus many of the more than two dozen chemicals used in fumigation. 

SR-123-02 The Need​le Trap Device as Sampler or Focusing Trap: Quantitative and Sensitivity Comparability of GC Injection Port Solvent and Thermal Desorption Analyte Introduction 

T. Juarez, Uniformed Services University of the Health Sciences, Bethesda, MD; S. Strating, Uniformed Services University of the Health Sciences, Bethesda, MD; M. Stevens, Uniformed Services University of the Health Sciences, Bethesda, MD; D. White, Uniformed Services University of the Health Sciences, Bethesda, MD; P. Smith, U. S. Department of Labor--OSHA, Health Response Team, Salt Lake Technical Center, Sandy, UT 

Objective: Determine if a needle trap device (NTD) when used as a sampler or gas chromatography (GC) thermal desorption focusing trap has comparable accuracy and sensitivity with standard solvent desorption methods. 

Methods: Air sample were collected as one-minute samples using a manually operated pump to collect 10mL gas phase samples of benzene, toluene and tetrachloroethylene vapors at their respective peak exposure standards. NTD air samples were compared to coconut shell charcoal tube samples. Analytical transfer studies were conducted by directly spiking sampling media with analytes and compared with liquid calibration standards. A commercially available person-portable thermal desorption module was used to accomplish the analytical transfer of analytes from a larger capacity (6.4mm x 89mm) thermal desorption sorbent tube to a NTD. To assess the effect of NTD injection port insertion depth on detector response, responses were measured at four insertion depths. A flame ionizing detector (FID) was used for detection and quantification. 

Results: Similarity of calibration curve slopes and coefficients of determination indicate comparable accuracy and sensitivities between standard solvent desorption and NTD thermal desorption within the GC injection port (R2 = 0.94-0.98). FID response differences based on NTD insertion depth were observed and ranged from 300-445 percent increases from 53mm insertion to full insertion depth. Sensitivities between air sample calibration using NTD and traditional liquid calibration were made comparable by the use of an internal standard. 

Conclusions: This preliminary work demonstrates a modular approach to sampling analysis that is a more efficient, safer and environmentally responsible approach than traditional analytical methods involving solvent desorption when assessing short-term and peak concentrations. Future work will employ a person-portable GC/MS to achieve near real-time results. 

SR-123-03 Elimination of Sensor Maintenance by Sensor Interrogation and Correction 

T. Scheffler, Mine Safety Appliances Co., Cranberry Township, PA 

Objective: A bump check is a recommended best practice for users of gas detection. Understanding that, our focus was to develop and apply methods of simulating bump checks of amperometric electrochemical sensors without applying the gas of interest, yet at the same time providing the user with a high degree of assurance that sensor/instrument combination is functioning properly. 

Methods: Development of this “gas-less” bump check required two major efforts. Long term (over a period of several years) observation of sensor behavior was necessary and was conducted over several years. This was coupled with accelerated aging techniques. Linear and non-linear regression analyses were applied to develop a model of sensor behavior over time or “aging”. An electronic sensor interrogation technique was developed. The statistical techniques mentioned above were used to demonstrate that these interrogation techniques correctly predicted the behavior of the sensors with regard to aging. Second, an electrochemical sensor was developed with two independent channels. The first, an analytical channel, was designed to detect the gas of interest, and the second, a non-analytical channel, responds to some component of exhaled breath. Numerical analysis was used to provide a quantitative measure of the efficiency of gaseous transport into the senor. 

Results: The two independent tests of sensor functionality were evaluated with regard to their accuracy and statistical robustness. The electronic interrogation of the analytical channel correctly identifies and corrects for three major effects: long-term sensor aging, mid-term environmental effects, and effects due to inhibitors or poisons. The exhalation test interrogates the flow state of all membranes and barriers between the external atmosphere and working electrode of the sensor. The techniques described provided the correct indication of instrument functionality with greater than 98 percent accuracy, verified over time and sensor type. 

Conclusions: The deployment of this “gas-less” bump check methodology will result in greater end user awareness of gas hazards, safety and implementation of recommended best practices. This is of particular benefit as the responsibility for daily or periodic bump checks are frequently in the hands of the end user. 

SR-123-04 Estimating Integrated VOC Exposure in Near Real-Time 

E. Floyd, University of Oklahoma, Oklahoma City, OK; C. Lungu, University of Alabama at Birmingham, Birmingham, AL 

Objective: Exposures to Volatile Organic Compounds are a persistent issue in Occupational Hygiene. Integrated sorbent based assessments are the gold standard for exposure measurement since real time devices lack chemical specificity and can be subject to interference. Unfortunately, analysis of integrated sorbent samples is slow and exposure data may lag weeks behind the sampling event. This may leave hygienists to use cautious professional judgment until the sample results return. Photothermal desorption (PTD) can be used to release a portion of an integrated sampler for near real time estimation of exposure while still allowing the same integrated sampler to be analyzed by conventional means. 

Methods: Single walled carbon nanotubes (SWNT) were suspended in a solvent by ultrasonication and deposited on a porous membrane creating isotropically aligned felts. The felts were desorbed and degased by thermal treatment at 200oC overnight. Felts were loaded into a diffusive sampling body and exposed to toluene at three concentrations (1, 5, 10 ppm) for varied times (15, 60, 120 min; n = 6 each) in a well-mixed exposure chamber. Additionally the chamber concentration was verified real-time with a calibrated PID and by duplicate active sampling in accordance with NIOSH method 1501. The exposed SWNT felts were then desorbed by PTD and quantitated with a field grade PID; the same exposed SWNT felts were then extracted in CS2 with an adapted NIOSH 1501 method and results compared to near-real time estimates computed from PTD. 

Results: PTD released approximately 1 percent of the initial mass from each sample. At low concentration and shorter time the variability in the exposure estimate was greater than at high concentration and longer exposure, however both were predictive of actual exposure. The sensitivity of the PTD / PID near real time detection system was sufficient to estimate exposure within 25 percent of the true mean with 95 percent confidence. 

Conclusions: Photothermal desorption of integrated samples and quantitation with PID has tremendous potential to provide near real time estimates of personal exposure to simple VOC matrices. When coupled with chromatographic tools such as a GC there exists potential to perform in-field analysis of complex VOC exposures with laboratory confirmation of the same field sample. ​