Detecting danger

Figure 1. Aerial footage of the explosion at a petrochemical storage facility in Deer Park, Texas.

Even after the fire was extinguished, toxic chemicals in the plume remained a legitimate threat to air quality levels – one particular concern was benzene. Elevated levels of the chemical in the air around Deer Park led city officials to issue shelter-in-place orders and district schools to cancel classes as well as all after-school activities.

According to the Centers for Disease Control, the toxic gas, which can be found at various levels in volcanic eruptions, forest fires, crude oil, and even cigarette smoke, can cause shortness of breath, headaches, dizziness, confusion, an irregular heartbeat, tremors, and unconsciousness.¹ Extreme or extended exposure can ultimately lead to death.

Moreover, because benzene is a carcinogenic even in low concentrations, the National Institute for Occupational Safety and Health (NIOSH) has established the Recommended Exposure Limit (REL) for benzene to be 0.1 ppm.² Since benzene carries such a slight REL, the chemical requires extremely low level detection – it often has to be measured at levels at or below 1 ppm.

However, gathering benzene measurements in the field, can be extremely difficult. One potential detection method is traditional air sampling coupled with laboratory analysis. This is not only expensive, but it is also too time consuming in an emergency.

Photo-ionisation detectors (PID), often used in combination with pre-tubes, are a well-proven technology for detecting volatile organic compounds (VOCs). However, while PIDs can detect benzene, they can simultaneously detect other VOCs as well. This severely limits their effectiveness for use by emergency response teams in the field. For starters, the presence of other VOCs, such as toluene and xylene, can cause a false positive reading, which can lead to false alarms and misinformation being distributed to impacted communities.

For PIDs to effectively isolate and accurately measure benzene gas, emergency response teams would need to be equipped with gas detector tubes, or pretubes. Pretubes are glass vials that are filled with a chemical reagent that reacts to a specific chemical or family of chemicals. They are therefore necessary for PIDs to precisely identify whether or not benzene is present in the air. However, pretubes are consumables with a limited shelf life and it is impractical, if not impossible, for emergency response teams to constantly maintain a full and up-to-date inventory of gas detector tubes. Furthermore, the larger the emergency, the less likely sufficient quantities of pretubes will be maintained in stocks.

Figure 2. Various molecules are separated from one another using GC. They exit the columns at different times.

New technology changes measurement methods

Just before the Deer Park incident, Dräger, an international manufacturer of medical and safety technology, introduced new technology that combines the PID sensor with another reliable technology, gas chromatography (GC).

The new technology makes it possible to target and quantify pre-defined VOCs in a matter of seconds. This means that a team can quickly and effectively test for one particular substance, such as benzene, significantly reducing the need to wade through measurement data on other benign compounds that might be present.

Moreover, this technology, together with an advanced, proprietary algorithm, can detect and measure more than 29 toxic gases to the ppb level, depending on the gas. Specifically, the technology can measure concentration levels as low as 50 – 70 ppb for benzene. This is critical when attempting to measure a gas like benzene, given its especially low REL.

How the combined technologies work

Dräger’s device combines two tested technologies – GC followed by PID – to successfully achieve selective VOC detection. GC uses a column to separate different gas molecules based on their boiling points. Different gas molecules remain on the column for a certain amount of time, known as the retention time. As ambient air passes through the column, gases with lower boiling points generally exit more quickly, while gases with higher boiling points take more time to exit. While traditional laboratory technology requires the use of compressed gases to transport samples through the instrument and to fuel a flame detector, the Dräger X-PID effectively utilises ambient air as a carrier gas and does not require compressed gas to fuel a flame detector.

The device employs photo-ionisation – gases exiting the GC column are ionised by UV light emitted in the PID. The high-energy light ejects electrons from specific molecules, creating an electrically conductive gas, called a cold plasma. If it hits the detector, a space between two wires becomes conductive and a current flows.

Separating the substances in the GC column and subsequently ionising and detecting them will create a characteristic model of a sample. Each detected gas is shown as a peak on a time scale. Highly volatile substances will be found at the start of this scale, with less volatile molecules following behind. Complex mathematical algorithms help separate peaks that are in close proximity or overlap and assign them to specific substances – by type and concentration – using databases.

Figure 3. Compounds exiting the GC column are ionised and registered in the sensor module.
This generates a response signal which is then used to evaluate how much of a compound is present.

The technology would be particularly advantageous in emergencies such as the one Deer Park experienced in 2019. When a variety of VOCs are present, as was the case in Deer Park, running conventional GC in a laboratory is slow and impractical. Each compound has to be separated, so a longer GC column must be used, sometimes reaching 60 m or longer in length and easily taking 10 min. or more to measure.

X-PID technology uses a shorter GC column while at the same time limiting the compounds to be analysed – this allows for benzene to be isolated and measured in as little as 30 sec., even if other VOCs like toluene and xylene are present. By minimising cross sensitivity compared to a standard gas detection PID, this new technology also significantly reduces the potential for a false-positive measurement and resulting false alarms.

Additionally, this technology does not require the use of pre-tubes, so it is faster and easier than conventional methods. Perhaps most importantly, there is no expiration date so the need to continuously replenish pre-tubes that have limited shelf lives is taken out of the equation.

Other features include a robust exterior design to ensure the device performs in the most severe conditions. In addition, a temperature controlled sensor unit maintains a constant temperature above the ambient air temperature and separates water vapour from the target compounds to ensure reliable measurements in harsh environments.

Figure 4. Evaluation: certain substances appear on a scale based on the time they emerged (retention time)
from the GC column and its signal response (corresponding to amount). Powerful algorithms help separate
overlapping curves and assign them to specific substances – by type and concentration – using databases.

Conclusion

No one ever wants to see or have to respond to a chemical plant fire, a refinery explosion, or an accidental release of toxic chemicals. However, in the petrochemical industry, the potential for such an event is a constant and unpleasant reality. Therefore, planning and preparation are essential – that includes identifying the best tools and technology to properly measure air quality and detect volatile compounds – so when emergencies do happen, company personnel, emergency response teams and community officials can be equipped to respond with accurate and exact data.

References

1. ‘Facts About Benzene’, Centers for Disease Control and Prevention, https://emergency.cdc.gov/agent/benzene/basics/facts.asp
2. ‘Occupational Safety and Health Guideline for Benzene’, National Institute for Occupational Safety and Health, https://www.cdc.gov/niosh/docs/81-123/pdfs/0049.pdf

Written by Mark Heuchert, Dräger, USA.

Read the article online at: https://www.hydrocarbonengineering.com/special-reports/22052020/detecting-danger/