Field-Deployable, Real-Time Sensor for the Monitoring of Volatile Organic Compounds (VOCs)

Volatile organic compounds (VOCs) play an important role in atmospheric chemistry. They are critically involved in aerosol formation, tropospheric chemistry, and air quality. Specifically, several key VOCs have been implicated in health-related issues ranging from immediate, minor irritants to long-term exposure carcinogens. Due to these issues, government environmental agencies routinely monitor VOCs at sites known to house potential toxins (e.g., Superfund sites).

Figure 1 – Field-deployable sensor for real-time monitoring of volatile organic compounds (VOCs) (Los Gatos Research).

Conventional VOC monitoring involves sampling ambient air in a chemically passivated container made from electropolished stainless steel (e.g., summa canisters). The sample is then transported to a laboratory for subsequent analysis by a gas chromatograph coupled to a mass spectrometer (GC/MS).1 This technology has several advantages, including the ability to accurately quantify multiple species with high sensitivity. However, conventional GC/MS requires discrete sample capture and transport, prohibiting both real-time VOC measurements in the field and the quantification of highly reactive gases. Moreover, because of the long time between sample collection and analysis, the method has extensive sample handling protocols and associated consumables (e.g., carrier and calibration gases, GC columns, and vacuum pump rebuilds).

The GC/MS usually requires a highly skilled, dedicated operator, further inflating operating costs. The recent development of field-deployable GC/MS analyzers has addressed some of the aforementioned limitations; however, the systems are still expensive, highly complex, and require extensive consumables.

Figure 2 – Cavity ringdown trace fit to a single exponential. The decay time constant is directly related to optical absorption in the cavity and can be used to calculate the gas concentration.
Figure 3 – By measuring optical loss as a function of wavelength, the VOC sensor produces a mid-infrared optical absorption spectrum. The spectrum is fit to a weighted sum of catalogued absorption spectra due to VOCs and other atmospheric constituents. Major contributions from H2O, CO2, trichloroethylene (TCE), and tetrachloroethylene (PCE) are labeled.

Other highly sensitive VOC monitoring methods include proton transfer reaction-mass spectrometry (PTR-MS)2 and selected ion flow tube-mass spectrometry (SIFT-MS).3 These techniques involve using mass spectrometry to directly quantify ambient VOCs. However, unlike conventional mass spectrometry, which uses electron impact ionization that can fragment VOCs, PTR-MS and SIFT-MS use “softer” ionization techniques that involve chemical ionization. These technologies have already resulted in promising results; however, the instruments retain the complexity, expense, maintenance, and skilled operation associated with mass spectrometry.

Fourier transform infrared (FTIR) spectroscopy has also been used to quantify high levels of VOCs.4 It has the ability to measure several compounds simultaneously and has been industrialized for continuous, real-time field deployment. However, FTIRs are insufficiently sensitive for ambient air VOC monitoring, thus limiting the use of this technology to source monitoring.

Mid-infrared spectrometer for real-time quantification of VOCs

In order to address the limitations of current monitoring technologies, Los Gatos Research (LGR) (Mountain View, CA) has developed an ultrasensitive mid-infrared spectrometer that can provide highly accurate, real-time quantification of VOCs in the field. A schematic of the instrument is shown in Figure 1.

The system consists of a widely tunable (7–12 μm), external cavity-quantum cascade laser (EC-QCL) coupled to a high-finesse cavity comprised of two highly reflective mirrors (R = 99.67–99.92% from 9.4 to 11.6 μm or R = 99.89–99.94% from 7.9 to 10.4 μm, depending on the target compounds). Light transmitting through the cavity is focused onto a highly amplified HgCdTe detector. In order to increase the coupled laser power and thus improve the instrument’s signal-to-noise ratio, the incident laser beam is reinjected into the cavity.5

Figure 4 – Measurements of TCE, PCE, benzene, ethylbenzene, p-xylene, and o-xylene in a background of zero air taken over 3–14 hours. Measurement average and precisions are indicated in each panel.
Figure 5 – Measurements of varying concentration of TCE and PCE showing the linear response of the instrument.
Figure 6 – Ambient air was doped with TCE levels ranging from 4.22 to 17.74 ppb. The measured values were accurate to within the instrument precision for all three TCE levels.

To mitigate interfering optical absorptions due to ambient water vapor, the sample is dried using a regenerative Nafion dryer. Autonomous three-way valves are included to allow for periodic measurements of baseline (i.e., zero air) and multiple measurement locations within a building.

The laser is operated in pulsed mode at 62.5 kHz. Each pulse is accompanied by a cavity ringdown6 decay similar to that shown in Figure 2.

The cavity ringdown trace can be fit to a single exponential decay, where the decay time constant, τ, is related to the mirror reflectivity, R, and optical absorption due to the gas sample. The latter is given by the product of the molecular cross-section (σ), cavity length (l), gas concentration (C), and speed of light (c):

The laser is then tuned over the mirror high reflectivity range (9.4–11.6 μm or 7.9–10.4 μm) in 4–14 min, and the optical loss as a function of wavelength is determined (Figure 3). The resulting optical absorption spectrum is similar to that measured by FTIR, but is far more sensitive due to the very long effective optical pathlengths (Leff > 300 m).

The optical absorption spectrum is fit to a weighted sum of catalogued absorption spectra7 of VOCs and other common atmospheric constituents (i.e., H2O, CO2, NH3, etc.). The catalog includes more than 400 compounds and the weighting factors can then be used to directly determine the concentration of each of the different constituents. The robustness of the analysis can be improved and the cross-interferences mitigated by limiting the fit compound library.

Figure 7 – Ambient air was doped with 10.83 ppb of PCE. The measured values were accurate to within the instrument precision.

The resulting instrument provides the accurate quantification of multiple VOCs and other atmospheric constituents every 4–14 min without any consumables or laborious sample conversion processes. By packaging the unit within a thermally controlled, rack-mountable enclosure, the VOC sensor can be field-deployed for continuous, real-time monitoring.

Laboratory testing

The analyzer was extensively laboratory tested to determine its precision on a variety of key VOCs, including trichloroethylene (TCE), tetrachloroethylene (PCE), benzene, ethylbenzene, p-xylene, and o-xylene. Each gas was measured independently in a background of zero air continuously for 3–14 hr. The resulting measurement precisions are shown in Figure 4. Additionally, the instrument exhibits a highly linear response to varying concentrations of VOCs. The measured concentration of TCE and PCE in zero air backgrounds over a range of 10–100 ppb is shown in Figure 5.

Figure 8 – Estimated Extent of TCE in Shallow Groundwater and Vapor Intrusion Study Area. (Figure 8 was reprinted from a U.S. EPA Record of Decision Amendment [RODA] and is reproduced with permission.)

VOC analyzer deployment

Subsequent to the laboratory validation studies, the VOC analyzer was deployed to measure ambient air outside LGR’s laboratories in Mountain View, CA. The outside air was found to contain insignificant levels of VOC contamination. Thus, in order to verify the instrument’s performance for ambient air monitoring, the outside air was doped with TCE levels ranging from 4.22 to 17.74 ppb. The results, shown in Figure 6, demonstrate that the VOC analyzer is accurate to within its measurement precision for all measured TCE levels.

Figure 9 – Measurements of TCE in a steam tunnel and the breathing zone at the MEW Superfund site.

The experiment was repeated for ambient air doped with 10.83 ppb PCE, and the instrument was once again found to be accurate to within its measurement precision (Figure 7).

The VOC analyzer was then deployed at the Middlefield-Ellis-Whisman (MEW) Superfund site in Mountain View, CA (Figure 8). There are several groundwater pollution sources in the area, including defunct dry cleaners, engine cleaning/rebuild shops, and semiconductor manufacturing facilities. Much of the area has been remediated or otherwise contained, but Building 10 in Moffett Field Naval Station was only recently provided with ventilation to reduce the vapor intrusion from steam tunnels supplying an adjacent hangar. Additionally, Building 10 has been extensively surveyed using summa canister sampling coupled with conventional GC/MS.

Figure 10 – Measurements of PCE in a steam tunnel and the breathing zone at the MEW Superfund site.

The VOC analyzer was deployed in Building 10 of Moffett Naval Air Station for 1.5 weeks, during which time it operated unattended, 24 hours per day. Daytime low and high temperatures ranged from 50 to 80 °F (10 to 27 °C). During the deployment, the instrument was configured to draw 3.3 SLPM of sample air in alternating order from the steam tunnel and the breathing zone. The system was tested both with and without the remediation fans installed. The deployment results are shown in Figures 9 and 10 for TCE and PCE, respectively.

The data clearly indicate that the remediation method dramatically reduces contamination within the steam tunnel. Likewise, they indicate that there is minimal migration of pollutants between the steam tunnel and the breathing zone. Finally, conventional analyses (U.S. EPA Method TO-15) were performed during the deployment period on August 8th. The results of these analyses are shown in the figures and indicate that the real-time VOC analyzer provided accurate results.

VOC monitoring: future work

The VOC sensor can be enhanced in several ways to improve its measurement precision and speed. Foremost, recently developed EC-QCLs provide higher power and faster, more stable tunability, resulting in a VOC analyzer that provides more frequent results with higher precision. Additionally, the sensor can be interfaced to a preconcentration system8 to improve the instrument’s sensitivity by a factor of 10–100. Finally, LGR will deploy the VOC monitor at a variety of other polluted sites to better study the emission and remediation of a wide range of volatile compounds.

References

  1. Compendium Method TO-15, Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specially Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS); U.S. EPA, Washington, DC, 1999.
  2. de Gouw, J.; Warneke, C. Measurement of volatile organic compounds in the earth’s atmosphere using proton-transfer-reaction mass spectrometry. Mass Spec. Rev.  2007, 26, 223–57.
  3. Smith, D.; Španěl, P. Ambient analysis of trace compounds in gaseous media by SIFT-MS. Analyst  2011, 136(10), 2009–32.
  4. Griffiths, P.R.; Shao, L. et al. Completely automated open-path FT-IR spectrometry. Anal. Bioanal. Chem.  2009, 393(1), 45–50.
  5. O’Keefe, A.; Gupta, M. et al. Absorption spectroscopy instrument with increased optical cavity power without resonant frequency build-up. U.S. Patent No. 7,468,797. 23; Dec 2008.
  6. O’Keefe, A.; Deacon, D.G. Cavity ring-down optical spectrometer for absorption measurements using pulsed laser sources. Rev. Sci. Instruments  1988, 59, 2544–52.
  7. Sharpe, S.W.; Johnson, T.J. et al. Gas phase databases for quantitative infrared spectroscopy. Appl. Spectrosc.  2004, 58, 1452–61.
  8. Dettmer, K.; Engewald, W. Adsorbent materials commonly used in air analysis for adsorptive enrichment and thermal desorption of volatile organic compounds. Anal. Bioanal. Chem.  2002, 373(6), 490–500.

Manish Gupta, Ph.D., is Chief Technology Officer, and J. Brian Leen, Ph.D., is Senior Scientist, Los Gatos Research, 67 East Evelyn Ave., Ste. 3, Mountain View, CA 94041, U.S.A.; tel.: 650-965-7772, ext. 226; e-mail: m.gupta@lgrinc.com.

Comments