Cadmium: finding the exposures nobody labeled as cadmium work

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.

Cadmium is unusual in one respect. A single result below the action level does not end your monitoring obligation, no matter which round of monitoring it came from.

OSHA's cadmium standard says a result below the action level must be confirmed by a second measurement taken at least seven days later, whether the result came from initial or periodic monitoring. Only then can monitoring stop.

Most of OSHA's other substance-specific standards handle this differently. The lead, benzene and formaldehyde standards all let an initial result below the action level end the obligation by itself, and their seven-day confirmation requirements apply to periodic monitoring only.

Silica and hexavalent chromium also end the obligation on an initial result below the action level, with no confirmation at all. Beryllium has a seven-day requirement, but again for periodic monitoring. Cadmium is the one that extends the requirement to initial monitoring.

A cadmium survey meant to end your monitoring obligation therefore takes two visits, or one visit with a return trip on the calendar. One day of sampling cannot end the obligation by itself.

Where we find it

• Cadmium electroplating. Plating and stripping tanks, rack loading and drag-out.
• Nickel-cadmium battery manufacturing and recycling. Plate pasting, sintering, assembly and breaking.
• Silver solder and brazing alloys. Filler metal that contains cadmium is still in use, and brazing puts it straight into fume. A maintenance shop with one brazing bench can have the highest cadmium exposure in the building.
• Torch cutting, grinding and welding on cadmium-plated steel. Fasteners, aircraft hardware and older structural parts.
• Pigment and stabilizer manufacturing and handling. Cadmium yellows, oranges and reds in plastics, glass and ceramics.
• Scrap and metal recycling. Mixed scrap that contains plated parts and nickel-cadmium cells.

Exposure limits

Both limits are in micrograms per cubic meter of air (µg/m³), and both are 8-hour time-weighted averages (TWA).

Everything in the monitoring provisions of the standard ties back to those two numbers. No separate short-term trigger drives the monitoring schedule, and the whole standard runs off the 8-hour TWA.

There is no quarterly step

Under the cadmium standard, you must monitor workers exposed at or above the action level at least every six months. For exposures above the PEL, the standard does not name an interval at all. It says to monitor often enough to make sure the right respirators are selected and to evaluate how well your controls work. Controls here means engineering controls, such as ventilation, and safer work practices.

Six months is the minimum, not a schedule. Suppose your exposures are above the PEL and changing, as they would be with a new plating line, a new alloy or a change to the ventilation. Monitoring every six months may then not be defensible, because it is not enough to do what the standard says the monitoring is for.

The familiar “quarterly above the PEL” schedule comes from the lead standard. It is not in this one.

→ How often OSHA requires air monitoring, by substance

How we sample for it

The analytical method named in the standard's appendix is not the one to use. Appendix E of the cadmium standard describes an atomic absorption method (OSHA ID-189, flame AAS / AAS-HGA), which OSHA has withdrawn. The current validated methods are the three in the table: OSHA 1006 and 5003, which use ICP-MS, and ID-125G, which uses ICP-AES.

The instrument matters. Flame atomic absorption cannot measure amounts as small as an ICP-MS instrument can, and the number you need to see clearly is an action level of 2.5 µg/m³.

If the lab result comes back as “less than” a value at or above 2.5, the survey has not shown that you are below the action level. It has shown only that the instrument could not see that low with the volume of air collected.

Before the pumps go on your workers, ask which instrument the lab will use and what the smallest amount it can measure is at your sample volume. That five-minute conversation decides whether the survey answers the question.

The other field issue is straightforward. Cadmium exposure in brazing and torch cutting is intermittent and follows the task. A full-shift TWA on a technician who brazed for 40 minutes is a real number, but it tells you nothing about how well your controls worked during those 40 minutes. Where the task is short and hot, the sampling plan should cover the task as well as the shift.

What your results require you to do

• At or above the action level: periodic monitoring at least every six months.
• Above the PEL: monitor often enough to make sure the right respirators are selected and to evaluate how well your controls work. Six months is the minimum, not the answer.
• To stop monitoring: a result below the action level, from initial or periodic monitoring, must be confirmed by a second sample taken at least seven days later.

The seven-day interval is a real scheduling constraint. Two samples on consecutive days do not meet it, and neither do two samples taken the same week on different workers.

When you may not need this

Your only contact with cadmium may be handling finished, sealed cadmium-plated parts. If there is no cutting, grinding, heating or abrasive contact, there may be no airborne cadmium to measure. We would rather tell you that after walking through your process than sell you a survey that measures nothing.

What we will not do is agree to that on the strength of a safety data sheet. Cadmium exposure hides in jobs nobody labeled as cadmium work: the brazing bench, the scrap torch, the grinder.

→ How many air samples do you really need?