Real-Time Detection Systems

PO123
Real-Time Detection Systems

Wednesday, June 3, 2015, 10:00 AM - 12:00 PM

CS-123-01 NIOSH Center for Direct Reading and Sensor Technologies: Enabling a New Era of Safety, Health, Well-being, and Productivity

D. DeBord, NIOSH, Cincinnati, OH; M. Hoover, Morgantown, WV

Situation/Problem: The use of sensors is expanding as countless remote wireless sensors are now employed for monitoring the environment, work sites, disaster relief, agriculture, and health. Newly developing wearable and even implantable sensors could aid in hazard and exposure assessment and control by allowing individuals to monitor their own environments. Smart phone technology has helped to drive this area. Little guidance, however, is available that can aid the occupational health and safety professional in selecting the sensor most appropriate for the situation at hand. Information on validation of newer sensors is not always available. Ethical concerns about the use of sensors must also be resolved.

Resolution: The NIOSH Center for Direct Reading and Sensor Technologies (NCDRST) was formed in May 2014 to coordinate a national research agenda on sensors and direct reading monitors; develop guidance on selection sensor validation, quality control, and training; and establish partnerships to collaborate in these activities.

Results: Through its NCDRST activities, partnerships, and collaborations, NIOSH intends to advance the development, validation, and application of these technologies to enable a new era of occupational safety, health, well-being, and productivity. The recent NIOSH evaluation of smart phone applications for assessment of noise is an example of work that can foster the use of advanced sensors and methods as primary tools for industrial hygiene in the 21st century. Building on the guidance for selection of gas and vapor technologies that was published in 2012, a chapter for the NIOSH Manual of Analytical Methods will be developed on the technologies, use, selection, and quality control issues.

Lessons Learned: In 2008, NIOSH, in partnership with AIHA, hosted a workshop on direct reading exposure assessment methods and monitors to identify gaps and needs in this area. Areas identified included basic research needs, standards and guidelines for the validation of new sensors, and development of training. Given the evolving nature of sensors, a collaborative and targeted effort is necessary to meet these needs. Additional information about the center and partnering opportunities can be found at www.cdc.gov/niosh/topics/drst/. 


SR-123-02 Field-portable Fourier Transform Infrared (FTIR) Spectroscopy for the Measurement of 1-bromopropane Concentrations in Air

P. Smith, OSHA, Sandy, UT

Objective: Measurement of airborne stressor concentrations in near real-time is desirable to obtain feedback on potential exposure intensity while still in the field. This allows optimization of traditional sampling, and also quickly provides information on how changes in production or the introduction of control measures may impact worker exposure. The objective of this study was to verify the usefulness of gas-phase Fourier transform infrared (FTIR) spectroscopy for measurement of airborne 1-bromopropane in a work environment.

Methods: A flow-through dynamic air standard generator was used with measured delivery of 1-bromopropane into ultra-high purity nitrogen to create quantitative reference spectra at concentrations ranging from 5.40 to 148 ppm. Field samples of approximately 15 liter volume were rapidly collected (<30 s) into 5-layer aluminized bags at a manufacturing plant where a degreaser was operated using 1-bromopropane. Approximately 2 liters of air from a bag were used in the field for FTIR analysis, and triplicate charcoal tube samples were then collected from each bag for lab analysis and comparison to the field FTIR results. Absorbance from 1350 to 950 cm-1 was used for FTIR quantitation of the target analyte. This region excludes strong absorbance related to carbon dioxide and water vapor, and includes peaks related to 1-bromopropane that in all cases demonstrated absorbance <0.5 absorbance units.

Results: The triplicate sample laboratory results were in good agreement for each bag that represented four field samples, and average 1-bromopropane concentrations of 6.2, 6.2, 12.7, and 25.3 ppm were measured in these bags. The matched FTIR results obtained in the field were 7.6, 6.9, 11.7, and 24.5 ppm 1-bromopropane respectively, and these data show that the gas-phase FTIR system results were highly correlated with those from charcoal tube sampling (R2 = 0.99).

Conclusions: The methods used to obtain quantitative reference spectra and then the use of the gas-phase FTIR system produced quantitative data that proved useful in estimating 1-bromopropane concentrations while still in the field. The ability to rapidly sample air into a bag, followed by quantitative analysis of that air in near real-time using gas-phase FTIR can be helpful in guiding traditional short-term sampling to cover periods of maximum exposure for analytes where this is relevant, and also to evaluate changes to processes and controls in near real-time.


SR-123-03 Real-Time Detection of ppb Levels of Benzene in Industrial Atmospheres

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

Objective: The OSHA method for benzene involves the collection on a charcoal tube and return to the lab, for desorption with carbon disulfide and analysis by gas chromatography (GC) with a flame ionization detector (FID). The target concentration for this benzene method is 1 ppm since that is the PEL. While the photoionization detector (PID) is an ideal screening tool for detecting low or even sub ppm levels of benzene, it responds to all VOC’s, so the measurement of benzene at or below 1 ppm is not very accurate. However, when we use a PID with a GC the detection for 1 ppm of benzene is precise. We propose to develop a hand-held GC with a PID to detect to detect low ppb levels of benzene in real time and “on site”.

Methods: The hand-held GC weighs less than 3 pounds, is able to operate for >6 hours on a battery, is easy to operate and has a small external tank for carrier gas (nitrogen) that can be clipped onto a belt. The hand held GC stores the peak height and concentration data. If a chromatogram is needed, up to 20 are stored in memory and if more are needed, the 0-1 VDC analog output can be connected directly to a PC with PeakWorksTM, the accompanying chromatography integration software.

Results: A permeation tube was used to generate ppb levels of benzene in air. A chromatogram of 500 ppb of benzene (OSHA action level) in air produces a large signal (0.5 VDC) on the GC. The detection limit for benzene is approximately 5 ppb or 1% of the action level. The analysis time is 2 minutes presently but with some optimization, it can be reduced to about 1 minute which would be ideal.

Conclusions: The GC/PID method provides a very rapid and sensitive technique for the real-time analysis of benzene at ppb levels in the field. The precision of benzene for 5 samples at 1/10th of the PEL was excellent at 5.8%. A special precolumn with back flushing was used to remove interferences.


CS-123-04 Worker Exposure Assessment Strategies for Tank Gauging Operations Conducted in the Bakken Oilfields (North Dakota)

J. Hill, P. Smith, D. Pearce, OSHA, Sandy, UT

Situation/Problem: The Bakken oil field encompasses a multilayered hydrocarbon deposit over nearly 200,000 square miles. Bakken crude oil has a light hydrocarbon profile, and hydrogen sulfide is often absent from the production stream. Across this field, workers manually gauge storage and production tanks, exposing them to gases and vapors emitted from the crude stream. Time-weighted average sampling is commonly used yet for gauging during Bakken crude production peak and IDLH exposure conditions may be more important.

Resolution: Several methods have demonstrated value in assessing tank gauging exposures at production sites, with focus on measurement of peak and/or IDLH exposures. Grab samples were rapidly collected into 5-layer aluminized bags, and two measurement approaches were then applied. Brief needle trap sampling was completed from a bag within minutes of air collection, and analysis was performed immediately using a person-portable gas chromatography-mass spectrometry (GC-MS) instrument to obtain hydrocarbon composition data. Triplicate sorbent tube samples were obtained from each bag for stabilization of grab sample analytes compatible with this sampling approach, and bulk crude oil samples were collected.

Results: Field GC-MS analyses avoided the potential loss of C3 and C4 analytes from bags and showed that these permanent gases were important components in the breathing zone exposure profiles, which were dominated by C3 - C6 hydrocarbons. Laboratory analyses of the bulk crude oil samples showed the presence of C3 - C24 compounds. While the sorbent tube stabilization method used did not measure analytes smaller than C5 due to lack of retention, at least one breathing zone grab sample stabilized on sorbent tubes contained combined C5 - C6 hydrocarbons at levels of about 1500 ppm.

Lessons Learned: This work demonstrates a need to determine the potential for IDLH conditions based on the sum of fractional IDLH values observed for compounds with similar physiological effects. Further work should be completed with sorbent tubes or whole-air analysis methods that include C3 and C4 compounds in the overall quantitative exposure assessment. Comprehensive analysis of whole-air samples may require the use of portable GC to avoid loss of very light analytes, and calibration of the person-portable GC-MS instrument used in this work as well as a larger van-mounted instrument is under investigation.


SR-123-05 Real-Time Detection System for Measuring VOCs, Temperature, Humidity, and Location

K. Brown, K. Mead, R. Kovein, P. Shaw, NIOSH, Cincinnati, OH; R. Voorhees, Univ. Cincinnati, Cincinnati, OH; A. Brandes, MeasureNet LLC, Cincinnati, OH

Objective: The objective of this project was to build and test a direct reading method (DRM) prototype instrument that contained an indoor real-time location system (RTLS). This personal exposure and location (PEAL) measurement device incorporated Wi-Fi data communication, radio location transmitters, and sensors for volatile organic compounds (VOC’s), humidity, and temperature. Antennas on the wall monitored the location device and sent real-time location data to a remote laptop. The sensor data was merged with location on work area floor plans and displayed on a laptop monitor.

Methods: The prototype PEAL measurement device was tested for accuracy and precision using two measurements at each of 4 locations and 3 gas concentrations with location and concentration randomized. Additional testing included a simulated chemical release scenario that resulted in 4 runs with 4 trials each. All test scenarios were conducted within an indoor 5000 sq. ft. work area. The simulated chemical was sampled 4 times once after each of the 4 runs. The calibration data were analyzed by regression analysis. The chemical release was confined to a room with elevated VOCs, temperature, and humidity.

Results: Location measurements were accurate to 1.5 meters with a standard deviation of 0.4 meters and obtained at a measurement frequency of 1.4 seconds. The PEAL’s PID sensor had accuracy of 9.9%. In other words, for a given fraction, say 0.95, and a given concentration θ, 95% of the measurements are expected to fall in the range (0.901xθ, 1.099xθ). The precision was 4.3% measured over a range of 0.2 to 10. ppm and with an LOD of 0.2 ppm. The PEAL measurement device was able to detect a simulated VOC release within seconds of entering an exposure area triggering an audio alert at both the PEAL device and the remote laptop.

Conclusions: This novel PEAL measurement device was able to instantaneously measure chemical vapors and wirelessly communicate their concentration and location to a remote lap top while providing personal alarms of exposure. The data were stored for later analysis, including a “four dimensional graph” (3 spatial dimensions and color for concentration) plot that produced an exposure hazard map. In practice, this 4-D map could subsequently be used to identify high hazard work locations and correlate them with a time and task. The PEAL device provided immediate feedback of exposure and later 4-D location-exposure map.


SR-123-06 Combatting Humidity and Other Matrix Gases on the Response Hand-held PIDs

W. Haag, Ion Science USA, Stafford, TX.

Objective: Hand-held PIDs have long suffered from 1) drifting high readings at very high RH, causing a false positive reading in the absence of VOCs and 2) quenching of VOC signal, causing a false low reading as RH increases. This paper discusses options for avoiding these effects, including humidity filtering tubes, Nafion humidity exchangers, RH compensation algorithms, and newer PIDs insensitive to humidity. This study also examined the effects of oxygen and methane on PID response.

Methods: Various manufacturers’ PIDs were tested by measuring the response at constant isobutylene concentration while varying the humidity from 0 to 95%. Gas mixtures were generated using a flow system to mix dry isobutylene with air humidified by passing through a water bubbler. PIDs with RH compensation were tested with compensation both on and off. Similarly, gas mixtures were generated at fixed VOC concentration while varying oxygen from 0 to 98% (in nitrogen) and with methane concentrations up to 20% in air.

Results: Instruments that incorporate a humidity sensor and attempt to compensate tend to show poor results. For example, a Tiger PID showed almost no humidity effect at all VOC concentrations, while a MiniRAE 3000 overcompensated by +60% at 10 ppm & 85% RH and undercompensated by -25% at 1000 ppm & 85% RH. Oxygen quenched the VOC response by as much as 40% low in pure oxygen. Between 0 and 40% oxygen, the VOC response was generally between +/-20% of the reading in air, and often much closer. Methane strongly quenches the VOC response when the methane concentration in about 1% by volume. Modern PIDs show less effect, but in pure methane the VOC signal is so strongly quenched in all cases that the readings are near zero.

Conclusions: 1) Recent developments in PID technology has resulted in PIDs with greatly reduced humidity effects. This avoids the need to use complicating features like humidity filtering tubes, Nafion gas humidifiers or firmware compensation algorithms. The simplest, most accurate solution is to use a newer PID that is inherently unaffected by humidity. 2) Variations in oxygen concentrations in calibration gases are not likely to affect calibration accuracy in modern hand-held PIDs. 3) Methane quenches the VOC response to a similar extent, although some PIDs are somewhat better than others. PIDs cannot be used to measure VOCs in natural gas.​