Solvent Exposure: Rebuilding a Shift TWA from Imperfect Samples
A flexographic press operator works 0700-1530 with an unpaid lunch from 1130 to 1200. You collect three personal samples across the shift: 2.5 hours at 62 ppm, 3.0 hours at 28 ppm, and 2.0 hours at 15 ppm of a solvent with an 8-hour PEL of 50 ppm and no ceiling limit. The lunch break and the final 30 minutes of the shift were not sampled. The pump used for the 62 ppm sample was calibrated at 2.05 L/min before sampling and 1.85 L/min after; the other two pumps drifted less than 2%. The occupational physician mentions two other press operators with recurring headaches, and a recent resin change increased the aromatic content of the solvent blend.
1. Calculate an 8-hour TWA using a conservative treatment of the unsampled time, and decide whether the exposure is in compliance. Then show how the flow drift on the first pump changes the number and your confidence in the result.
Read the worked answer
Sampled mass-time product
| Segment | Hours | Conc. (ppm) | ppm-hours |
|---|---|---|---|
| 1 | 2.5 | 62 | 155.0 |
| 2 | 3.0 | 28 | 84.0 |
| 3 | 2.0 | 15 | 30.0 |
| Total | 7.5 | - | 269.0 |
269 / 8 = 33.6 ppm if unsampled time is assumed zero. That assumption is not defensible, because unmeasured time is not clean time.
Conservative 8-hour TWA. Assign the 0.5 h of unsampled time the concentration of the nearest adjacent segment (15 ppm): (269 + 7.5) / 8 = 34.6 ppm. Assigning it 28 ppm gives 35.4 ppm; assigning it the highest observed 62 ppm gives 37.5 ppm. Every plausible treatment lands well below the 50 ppm PEL, so the exposure is in compliance on the sampled data.
Now correct for the drift. Concentration is mass divided by volume, and volume is inferred from flow x time. Pre-calibration 2.05 L/min, post 1.85 L/min: a 9.8% change, far beyond the +/-5% validity criterion used in most sampling methods. If the true mean flow was about 1.95 L/min, the sample volume was overstated by a factor of 2.05/1.95 = 1.051, so the concentration was understated: 62 x 1.051 = 65.2 ppm.
Recomputing: (65.2 x 2.5 + 84 + 30 + 7.5) / 8 = 284.5 / 8 = 35.6 ppm.
Judgement, not just arithmetic. The correct conclusion is not "35 ppm, compliant, closed." The first sample fails the calibration-stability criterion, so the defensible statement is that exposure lies in roughly the 34-40 ppm band, driven by an invalid sample. Before any compliance determination or control decision, resample the full shift with pumps that pass post-calibration, and capture the highest-exposure task explicitly.
2. Assume resampling confirms about 36 ppm. A supervisor argues the case is closed because the site is 'well under the PEL.' What is the defensible professional position, and what actions follow?
Read the worked answer
Compliance is a legal floor, not a health endpoint. Three facts break the supervisor's argument:
- 36 ppm is about 72% of the PEL - a band where a substance with irritant or CNS effects can produce symptoms in susceptible workers even though the legal limit is met.
- Two operators already report recurring headaches, which is effect data the exposure number does not explain. Symptoms under the limit mean the limit is not controlling the hazard for this population or this task mix.
- The resin change altered the aromatic content. The ppm figure was built for one analyte; the blend may now include co-exposures with additive or independent effects that were never sampled, and the PEL for a single component does not cover a mixture.
Actions, in hierarchy order:
- Re-characterize. Full-shift sampling that includes peak tasks, plus speciation of the new blend. A 2.5-hour average of 62 ppm can hide short excursions well above it; with no ceiling, check whether the substance has a short-term exposure limit or excursion limit you should be applying as good practice.
- Engineer the source. Local exhaust at the nip/ink tray and enclosure of the fountain beats dilution ventilation, which is sensitive to worker position and door openings. Do not accept PPE as the primary control when the source is capturable.
- Health and communication. Symptom questionnaire and follow-up for the two operators; update the SDS/hazard communication for the reformulated product; brief supervisors that 'under the PEL' is not a health determination.
- Document. Write the rationale so the next person does not reopen the case as settled.