LEL and Other Combustible Gas Concentration Units

We are all used to converting units from metric to standard and from US standard to UK standard (a UK gallon = 1.20 US Gallons) but gas concentrations take it to a new level of complexity. We have absolute concentrations: mg/m3, relative concentrations: % ppm, ppb, and flammable concentrations: % LEL and probably some others I have not thought of. This article will try to explain the differences between these units and why people use one over another.

A true concentration is the amount of gas present per unit volume. For this reason, many gas concentrations are given in units of mg/m3. While this unit is correct, and is widely used in government regulations and industrial hygiene, it is more common to use relative units such as parts per million (ppm) or parts per billion (ppb). Relative units such as ppm and percent have the advantage that they do not change with pressure.

Oxygen in the air is about 20.9 % by volume (for any volume of air, 20.9 % of it is oxygen). Room air is 20.9% and so is compressed air in a scuba diver’s tank at ~ 2,000 psi. Similarly if I want to calibrate a gas monitor with 10 ppm of for example carbon monoxide, I can purchase a cylinder of 10 ppm carbon monoxide at 2,000 psi and deliver it to my gas monitor at close to atmospheric pressure and it is still 10 ppm. If one were working in mg/m3, the concentration at 2000 psi would be about 135 times as high as at ambient pressure because for every unit of volume there is 135 times more gas present. Since these relative concentrations are by volume of gas, they are sometimes written as ppmv, ppbv or % v/v in order to distinguish other relative measures such as ppm in liquids which are normally by weight.

One difficulty that arises is that many toxic gas sensors respond to the concentration of gas rather than the relative concentration. Thus if a gas sensor is calibrated at sea level and then shipped to Denver, (5,280 ft), the pressure is about 80% of sea level and so the absolute concentration will of any gas component will be lower than at sea level for the same ppm or % volume. Users of ChemDAQ products will be pleased to know that altitude is not an issue for their equipment since there is an automatic compensation parameter set in the monitor at the time of installation to correct for differences in altitude between ChemDAQ (where the sensor modules are calibrated) and the end user. LEL sensors also respond to the concentration and depending on how the sensor is constructed, some oxygen sensors respond to the absolute concentration, and others to the relative concentration.

Most of us however, happily live our lives at relatively constant atmospheric pressure and so the two units are readily interconvertible (see free ChemDAQ ppm to mg/m3 converter on the ChemDAQ website [bottom of page]) and we can use relative units without problem.

Another reason for using relative units occurs with flammable gases where percent LEL is widely used. Most flammable gases have a concentration range over which mixtures with air will burn. If there is too little fuel then there is insufficient heat produce for the flame to propagate and similarly if there is too little oxygen, again there is too little heat for the flame to propagate. A typical gas such as propane has flammability/explosive limits in air of 2.2 to 9.5 % by volume. [Matheson Gas Data Book, 6th Ed.], so any propane/air mixture in this range is flammable and so is potentially explosive. The lower limit of flammability is called the Lower Explosive Limit (LEL) and the other limit is the Upper Explosive Limit (UEL). The LEL and UEL values vary from gas to gas.

If one working in an environment where there is a risk of an explosive gas mixture forming, then no-one really cares if the concentration is 2.1% volume, they want to know whether the atmosphere is explosive or not and whether they should get out. Flammable gas monitors for workplace safety are therefore calibrated in % LEL which provides an immediate measure of the risk of forming an explosive atmosphere. The gas monitors typically have a range of 0 to 100% LEL. If the concentration reaches 100% LEL, then there is a potentially explosive atmosphere and so the alarms are set lower, typically 10% and 20% LEL.

To convert from % LEL to ppm, it is necessary to know the Lower Explosive Limit. Using ethylene oxide (EtO) as an example, the LEL is 3% by volume, which is equal to 30,000 ppm. It is common therefore in facilities that use large amounts of EtO to have two different types of monitor. One set of monitors measures EtO at parts per million concentrations to warn about potentially toxic exposures (OSHA Permissible exposure limit for EtO is 1 ppm calculated as an 8 hr Time weighted average) the other set of monitors is to detect potentially flammable gas mixtures that threaten not only people but also the building. Hopefully before the LEL monitors go into alarm the ppm levels alarms will have warned everyone to clear the air, but the LEL monitors are used to increase ventilation or turn on the sprinkler system.

There are even more units of gas concentration out there which are less common, but if anyone wants more explanation of these, please leave a comment.

Using Hydrogen Peroxide Monitors to Measure Peracetic Acid Vapor

Peracetic acid (PAA) also known as peroxyacetic acid has become widely used as a disinfectant and sterilant in healthcare, food processing, meat and vegetable production, water treatment and many other industries. PAA is a strong oxidizer and a primary irritant and the health effects of over exposure, especially to the vapor are well known.

As a result of these risks the US-EPA has issued Acute Exposure Guidelines for PAA and there are three AEGL levels: “AEGL-1 is the airborne concentration, expressed as parts per million or milligrams per cubic meter (ppm or mg/m3) of a substance above which it is predicted that the general population, including susceptible individuals, could experience notable discomfort, irritation, or certain asymptomatic nonsensory effects. However, the effects are not disabling and are transient and reversible upon cessation of exposure.” AEGL 2 is the level where those exposed may experience “irreversible or other serious, long-lasting adverse health effects or an impaired ability to escape.” And AEGL 3 is the level where those exposed may experience “life-threatening health effects or death.”

PAA is normally found as an equilibrium mixture with acetic acid and hydrogen peroxide:

CH3COOH + H2O2 <=> CH3COOOH + H2O

Therefore, whenever PAA solution is used, in addition to PAA vapor there is also hydrogen peroxide vapor and acetic acid vapor, both of which have OSHA permissible exposure limits (PELs) of 1 ppm and 10 ppm respectively calculated as an eight hour time weighted average, both significantly higher than the AEGL 1 for PAA of 0.17 ppm. Currently there is no OSHA PEL for PAA or ACGIH TLV, though the ACGIH has proposed a 15 minute short term exposure limit (STEL) for PAA of 0.2 ppm. The analysis below will use the AEGL 1 but it is simple to adjust the numbers if the new ACGIH STEL is adopted.

ChemDAQ recently launched a PAA monitor as part of its Steri-Trac® gas monitoring system that also includes monitors for hydrogen peroxide. Prior to the launch of this product, there were no monitors for PAA and very few analytical methods available, despite the widespread use of PAA. Employers seeking to protect their workers would often rely on detecting only the hydrogen peroxide and acetic acid components, but this approach is flawed in that PAA vapor is more hazardous that either of the other two vapors and in mixtures with a high PAA content, it is the dominant vapor present.

The best strategy to designing a gas detection system is to assess which vapor presents the greater hazard and detect that one. As discussed above, PAA is usually used as an equilibrium mixture with hydrogen peroxide and acetic acid and it is supplied in a variety of blends, some with a high PAA/H2O2 ratio and some with a low PAA/H2O2 ratio. Ratios in commercial blends typically vary from 10:1 to 1:5 PAA:H2O2.

The last piece to the puzzle is the vapor pressures of hydrogen peroxide and PAA solutions. At room temperature the vapor pressures of PAA and hydrogen peroxide are 1.93 and 0.26 kPa respectively at 25 oC [CRC Handbook of Chemistry & Physics 76th Ed, Lange’s Handbook of Chemistry, 12th Ed). Combining the vapor pressure and the PEL/AEGL 1 for the two compounds gives the relative hazard (43:1 PAA/H2O2). Multiply this number by the ratio of PAA/H2O2 in the composition gives the risk factor for that PAA blend.

If we assume that the risk of a minority vapor can be ignored if the risk is less than 20 % for the combination of both PAA and hydrogen peroxide, then if the risk factor is less than 0.2, the predominant risk is hydrogen peroxide and a hydrogen peroxide monitor will suffice. If the risk factor is greater than 5, then the predominant risk is PAA and a PAA monitor only is sufficient. If the risk factor is between 0.2 and 5, then both PAA and hydrogen peroxide vapors pose a risk and both types of monitor should be employed.

The risk factors for several common blends are shown below. The risk factors for other blends may be readily calculated as described above.

PAA (wt %)	H2O2(wt %)	Risk Factor	Monitor
0.23	7	1.4	H2O2 and PAA
2	22	4	H2O and PAA
5	25	8.8	PAA
10	20	22	PAA
15	10	66	PAA
32	6	234	PAA

In Conclusion, for all of the brands and blends that we currently have data on (~ 70), none of them would be adequately monitored using a hydrogen peroxide monitor alone and those facilities using only a hydrogen peroxide monitor maybe seriously underestimating the exposure risk of their employees. The majority of PAA blends would require PAA monitors only and a few blends, with low PAA/H2O2 ratios, need both hydrogen peroxide and PAA monitors in order to adequately monitor the vapor.

Requirement to Monitor for Hydrogen Peroxide

One of the topics that we often receive questions on concerns whether there is a legal requirement to monitor for hydrogen peroxide. The short answer is that there is no regulation from OSHA explicitly saying that hydrogen peroxide must be monitored, just as there is no explicit order requirement to monitor for carbon monoxide in a steel mill or hydrogen sulfide in a petroleum plant.

The reason why there is no statement requiring monitoring, is because OSHA along with most other government agencies intentionally write their regulations to set goals not prescribe means to create a safe workplace. i.e. performance based versus prescription based regulation. the goal is a safe work environment, gas detection is a means to achieving it. There are two reasons for this performance based approach. The first is that developing regulations is a slow process and if OSHA were to specify a particular method, it would probably be obsolete even before the final rule was published in the Federal Register. The second reason is that the circumstances at every employer are different and so the means to solve an exposure problem at one facility may be inapplicable to another facility. For example, the same regulations governing workplace exposure to hydrogen peroxide apply to a hospital sterilizing medical equipment, a titanium plant using hydrogen peroxide to pickle titanium ingots to remove mill scale and a sewage treatment plant using hydrogen peroxide to reduce odor emissions.

The Occupational Safety and Health Act (1970), imposes a legal duty on employers to “furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees.”

[Sec. 5] The hazards of exposure to hydrogen peroxide vapor are well known and have been for decades, and OSHA sets the legal standard for when exposures to hydrogen peroxide are considered free from recognized hazards etc. in its Permissible Exposure Limits (PELs) “An employee's exposure to any substance in Table Z-1, the exposure limit of which is not preceded by a "C", shall not exceed the 8-hour Time Weighted Average given for that substance any 8-hour work shift of a 40-hour work week.” The permissible exposure limit for hydrogen peroxide is 1 ppm calculated as an 8 hr time weighted average and the employer has an affirmative legal duty to ensure that the PEL is not exceeded.

Many people believe that hydrogen peroxide is completely safe, after all it is sold in super markets for treatment of minor cuts. However, gas or vapor sterilization is achieved by exposing the articles to be sterilized to high enough concentrations of reactive gases or vapors to ensure that all microbial life is destroyed (probability survival < 1 in a million). If the concentration of hydrogen peroxide in the sterilizer is high enough to kill even bacterial in the sporoidal form, then in the event of a leak, the concentration is high enough to pose a risk to nearby workers.

Some people may have received assurances from the folks who sold them a hydrogen peroxide sterilizer that their equipment could never leak. People in sales are often very enthusiastic about their products and often portray them in their best light. Modern sterilizers available today are indeed designed and manufactured to the highest engineering standards, and most are tested for leaks as part of the design process. However, as with any complex piece of equipment components can fail, user error happens and of course wear and tear takes its toll. Even though the sterilizers contain many safety features and are designed not to leak, the manufacturers will usually acknowledge that leaks can sometimes occur. If you are assured that it cannot leak, just request a statement to that effect in writing.

ChemDAQ has many customers with monitors monitoring their hydrogen peroxide sterilizers. In case further evidence were needed that sterilizers can sometimes leak, last year, one of our hospital customers installed ChemDAQ’s gas monitoring system for their four new hydrogen peroxide sterilizers (no names here!). All four sterilizers emitted a cloud of around 20 to 40 ppm hydrogen peroxide each time the door was opened at the completion of the cycle, which would have been particularly harmful if people are reaching in to retrieve the load, especially since the NIOSH immediately dangerous to life and health level for hydrogen peroxide is only 75 ppm. The FDA’s MAUDE data base also provides other examples of sterilizer malfunction including exposure of workers to hydrogen peroxide vapor.

Employers must ensure that their employees are not exposed to hydrogen peroxide levels greater than the PEL, but hydrogen peroxide has almost no odor and so odor cannot be used to detect the presence of a hydrogen peroxide leak. Therefore absent some kind of monitor, it would be very difficult to measure the hydrogen peroxide concentration.

Some facilities use badges for hydrogen peroxide, but badges suffer from two major defects. A typical badge is worn for a shift and then sent to a lab to be analyzed (typically 1 - 2 weeks). Thus badges provide no warning of current exposure; they merely document exposures that have already happened. The second drawback is that leaks, like other faults usually occur at unexpected times and so if, for example badgering, is performed every month, then there will be between one to 31 days (plus badge analysis time) before any leak is discovered.

A continuous monitor offers greatly superior performance by providing the instantaneous hydrogen peroxide concentration, and alarms if the concentration goes too high thus providing real-time protection of employees. Most systems also include the capability to log data, calculate time weighted average exposures and warn if the OSHA PEL will be/has been exceeded and provide record keeping, reports etc. that enable an employer to demonstrate that their employees have not been exposed above the OSHA PEL.

In summary, installing a gas monitor for hydrogen peroxide is NOT mandated by OSHA, but OSHA does require that employers ensure that employees are not exposed to hydrogen peroxide over the PEL. Hydrogen is odorless and generally imperceptible until present at concentrations greater than the PEL and so some kind of analysis method is required to detect it. While the employer is free to employ any effective method to ensure that its employees are not overly exposed to hydrogen peroxide, continuous monitoring is the most effective method for employers to meet the OSHA requirement. Since hydrogen peroxide vapor is imperceptible until above safe levels, if there is a leak, Are You Safe? How do you Know?

Gas Stratification is Not Relevant to Gas Monitor Placement

We all know that objects more dense that water sink and those less dense than water float; and that light gases such as hydrogen rise and heavy vapors sink. In meteorology we see warm air rising over colder air masses and in science experiments we see denser gases like carbon dioxide being poured like liquids. If any more confirmation were needed, the disappearance of a helium filled balloon from a child's birthday party into the heavens should put all doubts to rest that lighter gases rise and heavier gases sink.

It therefore makes intuitive sense that a heavy gas will accumulate in low lying areas and lighter gases will collect in high areas. We often see published advice from gas detection vendors for example that sensors for ammonia (mol. wt. = 17 g/mol) which is lighter than air (av. mol. wt ~ 29 g/mol) should be placed up near the ceiling and monitors for heavier gases such as carbon dioxide (mol. wt. 44 g/mol) should be placed near floor level. Even though it makes intuitive sense, does gas stratification occur in practice?

Stratification is the extent to which the heavier gases tend to settle to the bottom and the lighter gases rise to the top of an initially uniform air mixture, in the absence of bulk air movement. In well ventilated areas, gas stratification is irrelevant. The air movement from the ventilation will mix up the room air sufficiently that the gas and vapor concentrations will be uniform with height above the ground. Therefore, in well ventilated work environments, such as a hospital sterile process department with a high air turnover (typically at least 10 air exchanges per hour), we recommend placing gas monitors for toxic gases about five feet of the ground so that they correspond to the breathing zone of individuals regardless of the identity or molecular mass of the gas or vapor being detected.

Stratification is widely believed by many to occur in locations where there is little air movement; however a 2009 paper by Badino, which discusses stratification of air in caves, provides a very good mathematical analysis which goes a long way to answering the stratification question. His analysis shows that stratification does occur, but it requires a column of static air several kilometers high to have a major impact; and so stratification will not be relevant to most occupational safety gas monitoring applications.

Caves full of deadly carbon dioxide do exist, as do other confined spaces such as sewers, storage vessels etc.; but Badino argues that these arise not because of stratification but because these gases and vapors form in the caves and diffuse out very slowly causing a local high concentration. He also points out that many of these situations are also dangerous because of the low oxygen concentration, which he argues is due to the oxygen being consumed in the reaction with organic matter rather than stratification. These confined spaces present a significant danger to anyone entering them, regardless of whether the mechanism is stratification, diffusion or another cause. Therefore, whenever entering a confined space, especially one with little air movement, it is important to follow the normal confined space entry procedures and regulations.

Theilacker and M. J. White conducted a study of gas diffusion and stratification after a helium leak at Fermi Lab. Since helium is such a light atom (mol. wt. = 4 g/mol) they had expected it to displace oxygen from ceiling, but they saw no difference in the readings of the oxygen monitors as a function of height above the floor. The conclusion of their studies with both helium and sulfur hexafluoride, a large heavy molecule (mol. wt. 146 g/mol) was that "modest gas velocities will fully mix the spilled gases with air. The gases remained fully mixed over long distances in tunnels, or for long times in enclosed spaces." In other words stratification is not a issue with gases under normal working conditions, even if your normal work environment is 25 feet underground in a four mile long particle accelerator tunnel.

While these two papers are not the end of the story, the take home message is the same as we have been saying for many years. In most work environments with good ventilation, stratification of gases is not going to occur to a significant extent. Thus if measuring the concentration of a lighter than air gas such as ammonia or a heavier than air gas such as ethylene oxide, for workplace safety applications, in both cases the monitors should be placed at head height or about 5' off the ground. However, confined spaces, especially those with little air movement, present real dangers, even if the cause is not gas stratification, and so normal confined space entry procedures and regulations should be followed.