Monitoring volatile organic compounds

Electronic air quality sensors have become cheap enough that putting one in a display case is a plausible idea. Whether it is a useful idea depends entirely on what question you are asking, because the sensors answer a much narrower question than their output suggests.

What a TVOC sensor actually measures

The common low-cost sensors are metal oxide semiconductor devices. A heated sensing layer changes resistance in the presence of reducing gases, and firmware converts that change into a number labelled total volatile organic compounds.

Three limitations follow directly from the physics. The sensor cannot distinguish between compounds, so a harmless alcohol and a corrosive organic acid both move the number, with different and unknown weightings. The calibration is against a reference gas that is not what is in your case. And the devices drift, are sensitive to humidity and temperature, and frequently apply an automatic baseline correction that quietly redefines whatever it has been sitting in as normal.

The result is a reading that has no established quantitative relationship to risk to an object.

What it is nonetheless good for

Detecting change. A stable reading that suddenly rises after new material was introduced is a genuine signal, even if the absolute value means nothing. This is the honest use case.

Comparing enclosures. Moving one sensor between several cases in turn, with adequate settling time, gives a relative ranking. That can be enough to identify which of six cases is the problem.

Confirming that something is decaying. A closed store of cellulose acetate film with a rising reading is telling you something real, though A-D indicator strips are the established tool for that specific job and are considerably more reliable.

The methods that produce numbers you can defend

Passive diffusion samplers. Small badges left in place for a defined period, then sent for laboratory analysis. They identify and quantify specific compounds, typically acetic acid, formic acid, formaldehyde and sulphur compounds. This is the right method when a number has to appear in a report, a specification or a dispute.

Metal coupon indicators. Small polished coupons of lead, silver and copper left inside an enclosure for months and inspected for corrosion. This is essentially a long, unaccelerated Oddy test in situ, and it has the great advantage of measuring the actual risk to the actual metals rather than a proxy. It is slow and qualitative, and it is often the most convincing evidence available.

The Oddy test itself, for the specific question of whether a material is safe to enclose. The Oddy test page covers the method.

Choosing by question

Question	Method
Is this material safe to put in a sealed case?	Oddy test, or published results
How much acetic acid is in this case?	Passive sampler, laboratory analysis
Is anything in this case corroding metal?	Metal coupon indicators in situ
Has something changed since last month?	Electronic sensor, watching the trend
Is my acetate film decaying?	A-D strips

If you deploy electronic sensors anyway

Log the trend rather than reading the instantaneous number. Disable automatic baseline correction if the device allows it, because otherwise a slowly worsening case will be silently normalised. Keep one sensor in a known clean reference location so you can distinguish a real rise from a general drift. And treat any alarm as a prompt to investigate with a better method, not as a finding in itself.