Hydrogen sulfide: ceilings, peaks and no 8-hour PEL

This page explains what the rules and common practice call for. Every site is different, and what a survey or a filing needs at your facility is something we work out with you after we see it.

Federal OSHA has no permissible exposure limit (PEL) for hydrogen sulfide in general industry that is set as an 8-hour time-weighted average (TWA). General industry covers most plants and fixed workplaces, so if your site falls under OSHA's general industry air contaminant limits, the number you were told to meet may not exist.

Hydrogen sulfide (H₂S) is listed in one of OSHA's air contaminant tables, Table Z-2. The limits in that table are not structured like the 8-hour TWAs most people are used to. For H₂S they work like this:

• Acceptable ceiling concentration: 20 parts per million (ppm). A ceiling limit is a concentration that must not be exceeded.
• Acceptable maximum peak above the ceiling: 50 ppm for 10 minutes, once only in an 8-hour shift, and only if no other measurable exposure occurs.

That is all there is for general industry. You cannot “meet the limit on an 8-hour average,” because there is no 8-hour TWA limit to meet.

Where the “10 ppm / 15 ppm” pair comes from

Search for the H₂S limit and you will keep finding “10 ppm average / 15 ppm short-term limit.” The pair is real, but it is not federal law for general industry.

Most of it is an old threshold limit value (TLV) from ACGIH, the professional body that publishes the TLVs. Before 2010, ACGIH paired a 10 ppm 8-hour TWA with a 15 ppm short-term exposure limit (STEL) over 15 minutes, and that is the main source of the numbers still in circulation. ACGIH has since lowered both, and its current limits are a 1 ppm TWA and a 5 ppm STEL.

The 10 ppm 8-hour TWA is also real for construction and shipyards. OSHA's construction and shipyard standards each set a 10 ppm 8-hour TWA, but those are different standards that cover different work.

The 15 ppm STEL is also a Cal/OSHA value. Cal/OSHA is California's state OSHA plan, and federal OSHA has no STEL for hydrogen sulfide.

So the honest answer to “what's the H₂S limit?” is: which of you is asking? A refinery turnaround crew, doing planned shutdown work under the construction standards, and the plant operators beside them can be under different numbers on the same site, on the same day.

The REL is NIOSH's recommended exposure limit.

Look at the last two rows together. The ACGIH 8-hour TWA is 1 ppm, and the level that is immediately dangerous to life or health is 100 ppm. A health-based long-term limit and immediate danger are therefore a factor of 100 apart, and the OSHA general industry ceiling sits at one fifth of the immediate danger level.

Where we find it

• Oil and gas production and sour crude handling. Wellheads, separators, tank gauging and produced water.
• Refining. Sour water strippers, amine units, turnarounds and vessel entry.
• Wastewater collection and treatment. Sewers, lift stations, headworks and digesters, plus the manhole nobody classified as a confined space.
• Manure pits and farm slurry handling. Agitating the pit is the critical moment, because it releases in minutes what accumulated over months.
• Kraft pulp mills. Recovery, digesters and the lime kiln area.
• Tanneries and hide processing.
• Landfills, and construction and demolition waste that contains gypsum.
• Any confined space with decaying organic matter and a low point.

How we sample for it

The direct-reading method (OSHA 1028) has a restriction that matters more than it appears to. Do not use it where dimethyl disulfide, dimethyl sulfide or mercaptans are present.

Read that against the list of sources above. Sour gas, refinery sour water, kraft pulping and sewers all routinely contain sulfur compounds of this kind. Mercaptans are the odorant added to natural gas, and kraft mills produce dimethyl sulfide as a normal part of the process.

The direct-reading method is therefore ruled out in a large share of the places where H₂S is most likely to be present. Sorbent tube sampling (OSHA 1008) is the fallback, and it has to be planned into the visit, not discovered on site.

This also affects personal gas monitors. They are essential life-safety equipment and workers should absolutely wear them, but in those settings they are not automatically the source of your exposure record. Life safety and exposure documentation are two different jobs, and they need two different tools.

The ceiling limit changes the survey

A full-shift average cannot demonstrate compliance with a ceiling limit, or with a 10-minute peak allowance. An H₂S survey is therefore not “hang a pump on someone for eight hours.” It is short-term sampling targeted at the moments when the concentration rises, which include:

• Tank gauging
• Sample collection
• Opening a manhole
• Agitating a pit
• Vessel entry
• Pigging (running a cleaning tool through a pipeline)

Note the condition built into the Table Z-2 peak provision as well. The 50 ppm allowance applies only if no other measurable exposure occurs during the shift. It is not a routine allowance, and a site with steady low-level H₂S plus a daily gauging peak does not get to use it.

→ How many air samples do you actually need?

Do not rely on your nose

H₂S has a strong odor at low concentrations. At high concentrations it deadens the sense of smell, an effect called olfactory fatigue, so the odor disappears as the concentration becomes dangerous. Workers describe this as “it cleared up.”

That is how a large share of H₂S fatalities happen, including the ones where a second person goes in after the first.

Any H₂S program built on “we'd smell it” is built on the most dangerous property of the gas. That is worth saying plainly on a page that is otherwise about sampling, because the sampling is the smaller part of what makes an H₂S site safe.

→ What to do when results exceed the PEL