Peracetic Acid: Parts Per Million in Water and in Air

One of the most common sources of confusion when talking to people about workplace exposure limits is the difference between liquid ppm and gas phase ppm. In both cases ppm stands for Parts Per Million.

Using Peracetic Acid (PAA) as an example, PAA is commonly used in dilute solution e.g. ~0.2% by weight solution. This means that for every 1000g or solution, 2g of it is PAA, the rest is water, hydrogen peroxide, acetic acid, surfactants etc. The ppm is just a fractional weight, similar to a percentage (parts per hundred).

For a gas or vapor, ppm is still parts per million, but now it is parts per million by volume because we do not normally deal with the weights of gases (except perhaps atmospheric pressure, ~14 psi, is the weight of the atmosphere on us and a liter of air weights about 1.2g). Instead we normally deal with gas pressures and volumes. The pressure of a gas mixture is the sum of the pressure of its components, at least to a reasonable approximation (Ideal Gas Law) and so it is more convenient to work in parts per million volume than weight.

The key message therefore is that a ppm gas is not the same as a ppm in liquid and so again using PAA as an example, while the EPA Acute Exposure Guide Line for PAA vapor is 0.17 ppm (AEGL 1, 10 min to 8 hr time weighted average); it does NOT mean that 2,000 ppm solution (0.2%) solution is immediately deadly. It is possible to estimate the gas or vapor ppm from a liquid ppm, but it is a little involved (To avoid details, skip to last two paragraphs).

A gas and vapor are very similar, in that both relate to chemicals in the gas state, but a vapor is a chemical whose liquid is below its boiling point. Thus if water evaporates from a glass of water, that is water vapor. If the water and air is heated above the boiling point of water (100 ^oC), then the water would be present as a gas.

Some liquids evaporate easily (alcohol for example), others barely evaporate at all (olive oil). There is an equilibrium between the liquid state and the vapor state known as the vapor pressure and as the temperature of a liquid rises towards the boiling point the vapor pressure increases until at the boiling point the vapor pressure equals atmospheric pressure. Vapor pressures for many compounds have been measured and tabulated in chemical handbooks and so values are readily available.

If the liquid is a mixture, such as PAA solution, then the vapor pressure of a component is proportional to the mole fraction of that component (fractional number of molecules of the component compared to all molecules, see Raoult’s Law).

If we know the vapor pressure of our component (Vapor Pressure of PAA = 1.93 kPa at 25^oC, CRC Handbook of Chemistry and Physics, 76th Ed, p 6-80), we can calculate the PAA vapor pressure of various PAA solutions. Similar calculations can be be done for the other components such as hydrogen peroxide as well.

PAA Concentrations

        Liquid (%, ppm)                     Vapor(ppm)

• 0.05%   500 ppm                   2.5 ppm
• 0.1%       1,000 ppm               4.9 ppm
• 0.2%      2,000 ppm               9.9 ppm
• 0.5%       5,000 ppm              25 ppm
• 1%          10,000 ppm              50 ppm
• 5%>        50,000 ppm            260 ppm
• 10%         100,000 ppm         540 ppm

These PAA vapor concentrations are in ppm. The estimates also depend on the concentrations of other components in the mixture such as acetic acid and hydrogen peroxide (always found in PAA solutions) since they affect the mole fraction calculation discussed above. For the estimates in the table above the acetic acid and hydrogen peroxide concentrations were 5 and 10% w/w respectively.

These estimates though are only approximations. In particular, they represent equilibrium values which are not found in most applications where PAA is used. For example if the PAA vapor is being swept away by a ventilation system then the vapor concentrations will not reach the levels in the table above. These vapor pressure calculations are useful estimating of the potential concentration of PAA vapor and this allow a rational basis for determining the risk of exposure to PAA and what means should be employed to keep workers safe.

Note: if the PAA is being sprayed, then the concentrations included aerosols or the complete evaporation of droplets in which case the vapor pressure calculations do not apply.

If the PAA vapor concentration has the potential to exceed safe levels, then the PAA vapor should be monitored. Continuous monitors for PAA and many other compounds with potentially hazardous vapors are readily available. Even if the PAA is controlled with ventilation or is used within dedicated equipment, the potential exists for it escape into the environment. Any equipment can fail from wear and tear, mechanical failure or user error. Even though there is no OSHA PEL for PAA, the EPA has issued Acute Exposure Guidelines for PAA as discussed above, the ACGIH is considering a 15 minute short term exposure limit of 0.4 ppm and even manufacturers of PAA such as Solvay recommend a TWA exposure limit of 0.2 ppm.

In summary, there is a lot of confusion between ppm vapor concentrations and liquid concentrations for compounds like PAA. The two are different but are related by a somewhat involved vapor pressure calculation. The calculated vapor pressures though are only estimates but are useful in determining if there is a risk of over exposure. If there is a significant risk of over exposure, then continuous monitors for PAA should be employed.

Hydrogen Peroxide Emission Problems from Sterilizers

The previous ChemDAQ blog discussed continued off-gassing of plastic parts sterilized in a hydrogen peroxide sterilizer as reported by Rika Yoshida, Hiroyoshi Kobayashi at a recent conference of the World Formum for Sterile Hospital Supply. This presentation also discussed hydrogen peroxide vapor emissions from hydrogen peroxide sterilizers. The sterilizers included several models of Sterrad^® sterilizer from Advanced Sterilization Products, and the V-Pro1^TM from Steris. The authors have also published some of these results in the Japanese Journal of Environmental Infections, and a full text copy of their paper is available.

The investigators measured the hydrogen peroxide concentration inside a sterilizer immediately after the end of the cycle, when people would be reaching in to remove the load, and found very high concentrations (34 ppm Sterrad 100S; 60 ppm Sterrad 200; and 13 ppm V-Pro1) in some cases close to the NIOSH Immediately Dangerous to Life and Health value of 75 ppm. Since the time to unload a sterilizer is fairly short, a single exposure will probably not exceed the OSHA PEL of 1 ppm calculated as a time weighted average over 8 hours, but it would probably exceed the 3 ppm 15 minute short term exposure limit found in some states (Washington and Hawaii) and even the OSHA PEL may be reached in a busy facility for someone running multiple loads a day.

ChemDAQ has received many reports from users of hydrogen peroxide emissions when the sterilizer door is opened at the end of a cycle. In a typical ChemDAQ installation the sensor is placed on top of the sterilizer and so the concentration measured will be much lower than would measured from a sensor placed inside the sterilizer chamber because the vapor gets diluted by the time it reaches the sensor.

Users often see small increases in hydrogen peroxide concentration, usually less than 1 ppm, though one model of sterilizer was found to emit much higher concentrations (~30 ppm) each time the door was opened. In this case, the hospital had four sterilizers, all new, all showing the same behavior. The manufacturer was unable to solve the problem and the hospital now instructs their technicians to open the sterilizer door at the end of the cycle and leave the area until the ChemDAQ monitor shows that it is safe to return.

Rika Yoshida, Hiroyoshi Kobayashi began their investigation in response to complaints of eye and respiratory system irritation from healthcare workers using hydrogen peroxide sterilizers. There are many similar reports of eye and respiratory system irritation from hydrogen peroxide sterilizers in the FDA’s MAUDE database and so it is likely that the results reported in their paper are not unique.

For many years ChemDAQ has been pointing out that all sterilant chemicals are potentially hazardous since they are designed to kill all microorganisms and therefore we recommend that all sterilant gases and vapors should be monitored.

Plastic Parts Off-Gas Hydrogen Peroxide Vapor After Sterilization

Ethylene oxide (EtO) as a sterilant gas has two main drawbacks; the first is health – EtO is an irritant and a carcinogen and the second is economic; a typical EtO sterilization cycle may take 12 to 15 hours. While the actual sterilization step may only last a couple hours, the rest of the time is needed for the sterilized product to aerate and give up the EtO trapped within it.

Since the mid 1990s many alternative gases and vapors have been explored as low temperature gas or vapor sterilants. Obviously none of them are harmless since sterilization is achieved by exposing the device to high concentrations of reactive gases. Typical gases used are hydrogen peroxide or ozone, both strong oxidants, and formaldehyde (steam formaldehyde), an alkylating agent similar to EtO (not currently used in the US).

In addition to avoiding EtO, the main selling features of hydrogen peroxide sterilization have been the short cycle times. The two leading manufacturers of hydrogen peroxide sterilizers, Advanced Sterilization Products and Steris Corporation, both now offer cycles times as short as 28 minutes. These features have been so attractive that hydrogen peroxide sterilization is now the dominant form of low temperature gas sterilization used in the US today.

The conventional wisdom is that EtO has a long cycle time because porous and some plastic products dissolve or otherwise retain the EtO and the long cycle time is required to allow the EtO to diffuse out of the product; whereas for hydrogen peroxide little or no aeration time is required.

A recent study by Japanese researchers from the Division of Infection Prevention and Control, Tokyo Healthcare University Postgraduate School challenged the conventional wisdom by showing that some plastic devices after hydrogen peroxide sterilization, off-gas the hydrogen peroxide vapor and take a much longer time for the vapor to clear than some of the current hydrogen peroxide sterilization cycles allow.

The researches took 11 kinds of plastic test panels, made of the polymers common used to manufacture medical devices and sterilized them using a hydrogen peroxide sterilizer and then measured the hydrogen peroxide vapor off-gassing from the surface of the plastics. They found that some plastics retained the hydrogen peroxide for much longer than others. A similar effect is seen with EtO, where some polymers such as PVC are notorious for slowly releasing dissolved EtO. The initial concentration of hydrogen peroxide emitted ranged from ~40 to over 300 ppm and many of the plastic panels with the initially higher hydrogen peroxide continued to emit concentrations over 50 ppm more than 25 minutes later.

In another test, the researches took a medical stapler made of polyetherimide, sterilized it, and found that the stapler initially emitted over 300 ppm hydrogen peroxide and took six days for the emitted hydrogen peroxide concentration to fall to 10 ppm, and 24 days before the concentration fell below 1 ppm (the OSHA PEL). In another test, flexible scopes continued to out gas hydrogen peroxide above 10 ppm for 18 to 40 hours.

The researchers started the work in response to complaints or eye and respiratory system irritation from healthcare workers who work around the sterilizers but the use, clean & sterilize and reuse times of many devices such as flexible scopes is often much less than 40 hours and so the researchers went to comment that healthcare workers should be on the look out for adverse effects in patients (as well as in their colleagues performing the sterilization).

The conventional wisdom is that hydrogen peroxide sterilization is much superior to EtO sterilization in large part because the latter does not require long aeration periods and so can perform its sterilization function with short cycles which saves time and resources as equipment can be sterilized and put back into service more quickly. This study by Rika Yoshida, Hiroyoshi Kobayashi directly challenges the belief that there is no significant out-gassing with hydrogen peroxide and shows that the concentrations being emitted are not only above the OSHA PEL (1 ppm 8 hr TWA), but in some cases over the NIOSH Immediately Dangerous to Life and Health level (75 ppm). These results parallel a study from 1997 by MacNeal and Glaser ["Comparison of healthcare based sterilization technologies: Safety, efficacy, and economics" in Journal of Healthcare Safety, Compliance & Infection Control (1997), 1,(2, December), p 91 to 107] where they found that

"Packages removed from the sterilizer after on hour continued to emit residual H2O2 gas at short-term or instantaneous concentrations of up to 2.5 ppm, for up to 1.3 hours following their removal from ther sterilizer."

Surprisingly, there do not appear to have been any other studies that looked at out-gassing of hydrogen peroxide from sterilized medical devices, though with the questions raised by these two research groups, we are sure that much more attention will be given to this important subject going forward.