Assessing silica exposure

Before you can choose the right controls, you need to know how much silica workers are breathing, or at least whether a recognized method of control will keep exposure low enough. This lesson explains the main ways exposure is assessed: personal air sampling, objective data, task-based approaches and direct-reading monitors. It covers how a sample is collected and analyzed, how to calculate and interpret results and how to plan an assessment for a group of workers. HSE-240 introduced exposure assessment for airborne contaminants in general; this lesson focuses on what is particular to silica.

Why assessment matters

Assessment answers three questions: are workers exposed above the limit or action level, which tasks drive the exposure, and are the controls working? Without it, employers either under-protect workers or spend money on controls and respirators that are not needed. Assessment is also a legal requirement in most jurisdictions, as Lesson 3 explained, and the results determine what else must happen, from written plans to medical surveillance.

Options for assessing exposure

Canadian regulations generally require the employer to assess exposure, but leave the method to professional judgement. In practice, four approaches are used, often together.

Personal air sampling

Personal sampling is the reference method because it measures what the worker actually breathes. A small battery-powered pump draws air through a sampler clipped to the worker's collar, within about 30 centimetres of the nose and mouth, the breathing zone. The sampler has two parts: a cyclone that spins out the larger particles, letting only the respirable fraction pass, and a filter cassette that collects the respirable dust.

Each cyclone is designed to operate at a specific flow rate so that its cut point matches the respirable convention. The flow rate must be set and checked with a calibrator before and after sampling. Using the wrong flow rate changes the cut point and biases the result.

Higher flow rates collect more air and therefore more silica, which matters because the amount of silica on a filter from a well-controlled task can be very small. Laboratories analyze the filter by X-ray diffraction (for example NIOSH Method 7500 or OSHA Method ID-142) or by infrared spectroscopy (for example NIOSH Method 7602). Results come back as the mass of quartz and cristobalite on the filter, usually in micrograms.

Calculating the concentration

The concentration is the mass of silica collected divided by the volume of air sampled:

Here m is the mass of silica on the filter, Q is the average flow rate and t is the sampling time. With mass in micrograms and volume in cubic metres, the result is in µg/m³. Since 1 m³ is 1,000 L, a flow rate of 2.5 L/min over 480 minutes samples 1,200 L, or 1.2 m³.

def concentration(mass_ug, flow_lpm, minutes):
    volume_m3 = flow_lpm * minutes / 1000
    return mass_ug / volume_m3, volume_m3

samples = [
    ("cut-off saw, wet", 9.0, 2.5, 455),
    ("tuckpointing grinder with shroud", 31.0, 2.5, 470),
    ("floor grinder, no vacuum", 210.0, 2.5, 430),
    ("helper, general clean-up", 6.0, 2.5, 480),
]

for task, m, q, t in samples:
    c, v = concentration(m, q, t)
    print(f"{task:34s} {m:6.1f} ug on {v:.2f} m3 = {c:6.1f} ug/m3")

cut-off saw, wet                      9.0 ug on 1.14 m3 =    7.9 ug/m3
tuckpointing grinder with shroud     31.0 ug on 1.18 m3 =   26.4 ug/m3
floor grinder, no vacuum            210.0 ug on 1.07 m3 =  195.3 ug/m3
helper, general clean-up              6.0 ug on 1.20 m3 =    5.0 ug/m3

These are concentrations over the sampled period only. To compare with an 8-hour limit, the result must be expressed as an 8-hour TWA, which depends on what happened during any unsampled time.

From sample to 8-hour TWA

Ideally, samples cover the whole shift. When a worker is sampled for only part of the shift, or several consecutive samples are taken, the 8-hour TWA is calculated by summing concentration multiplied by time and dividing by 8 hours (480 minutes):

Unsampled time must be assigned a value. If the worker was doing the same task, the sampled concentration may be assumed. If the worker was doing no silica work, zero may be reasonable. If you do not know, assume the worst case or sample longer.

Detection limits and very low results

Laboratories cannot measure arbitrarily small amounts of silica. The limit of detection for quartz by X-ray diffraction is typically a few micrograms per filter. Dividing by the air volume gives the lowest concentration the sample can detect. Short samples or low flow rates can produce detection limits close to, or even above, the action level, which makes the result useless for compliance.

lod_ug = 5.0   # illustrative limit of detection per filter, micrograms

print("minimum detectable concentration (ug/m3)")
print("flow   60 min  120 min  240 min  480 min")
for q in (1.7, 2.5, 4.2):
    row = []
    for t in (60, 120, 240, 480):
        v = q * t / 1000
        row.append(f"{lod_ug / v:7.1f}")
    print(f"{q:4.1f} " + " ".join(row))

minimum detectable concentration (ug/m3)
flow   60 min  120 min  240 min  480 min
 1.7    49.0    24.5    12.3     6.1
 2.5    33.3    16.7     8.3     4.2
 4.2    19.8     9.9     5.0     2.5

A 1-hour sample at 1.7 L/min cannot detect anything below about 49 µg/m³, so a result of 'not detected' says almost nothing about compliance with a 25 µg/m³ limit. Full-shift samples at higher flow rates give far more useful information. When a result is below the detection limit, report it as less than that value, not as zero.

Planning a sampling strategy

It is rarely practical to sample everyone on every shift. Hygienists group workers into similar exposure groups (SEGs): workers who do similar tasks with similar materials and controls and so are expected to have similar exposures. A few workers from each group are sampled on several days, ideally including the higher-exposure days.

Exposures vary a lot from day to day, often by a factor of two or more, because of weather, materials, production and how carefully controls are used. A single sample is a snapshot. Several results allow you to see the spread and judge how often the group might exceed the limit. A common rule of thumb is that if any result is above half the limit, the group needs closer attention.

Task-based tools in Canada

Several Canadian regulators provide task-based guidance for construction. WorkSafeBC's online silica control tool lets employers choose a task, material and control and see predicted exposure ranges based on a large database of measurements, along with the respirator needed. Ontario's ministry guidance on silica on construction projects groups tasks by expected exposure and lists controls and respirators for each. These tools are a starting point: they rely on controls being set up and used exactly as described, and they do not replace sampling where tasks or materials are unusual.

Where sampling is done, results should be compared with the limit that applies, and repeat sampling should follow when results are near or above it, when tasks or controls change, or at intervals set out in the exposure control plan.

Direct-reading monitors

Real-time dust monitors, usually light-scattering photometers, show concentration second by second. They cannot distinguish silica from other dust and need correction factors to estimate mass, so they are not used to show compliance. But they are excellent for finding which step in a task produces the most dust, comparing tools and checking whether a vacuum shroud or water feed is working. Some newer systems combine real-time monitoring with end-of-shift silica analysis on the same filter, giving faster feedback.

Interpreting and communicating results

Results should be shared with the workers sampled and with the joint health and safety committee or representative, in plain language. Explain what was sampled, the result compared with the limit, what it means and what will change. Workers have a right to see their exposure records. Keep records for as long as your regulator requires, which for carcinogens is often decades, because they may be needed if a worker later develops disease.