Air Quality Monitor Sensor Technology: PM2.5 Laser Scattering, NDIR CO2, and MOS VOC Accuracy Compared

Volume I  ·  July 2026  ·  4,601 words

A consumer air quality monitor is a compact electronic instrument that measures, in real time or near-real time, the concentration of airborne pollutants inside a building. The pollutants it measures — particulate matter, carbon dioxide, and volatile organic compounds — are chemically and physically unrelated, and no single sensor technology can detect all of them. A typical monitor contains three independent sensor modules housed in a shared enclosure with a display and a wireless radio. The quality of the data it produces is determined almost entirely by the sensor technologies selected for each of those three modules, the calibration applied at the factory, and the compensation algorithms that correct for temperature, humidity, and sensor drift over time. This analysis examines the operating principles of the three dominant consumer-grade sensor technologies, their accuracy relative to reference-grade instruments, and the practical considerations that determine whether a given monitor is producing meaningful data or a display of numbers whose relationship to the actual airborne pollutant concentrations is approximate at best.

The Three Pollutants and Why They Require Three Different Sensors

Particulate matter (PM), measured as PM2.5 and PM10, consists of solid particles and liquid droplets suspended in air. PM2.5 — particles with an aerodynamic diameter of 2.5 microns or less — penetrates deep into the pulmonary alveoli and crosses into the bloodstream; the World Health Organization estimates that ambient PM2.5 exposure contributes to approximately 4.2 million premature deaths annually worldwide, and the WHO's 2021 air quality guideline sets a 24-hour mean target of 15 µg/m³. Indoor PM2.5 sources include cooking, candle burning, tobacco smoke, fireplace operation, and infiltration of outdoor pollution through the building envelope. A particle sensor measures these aerosols but cannot distinguish a 2.5-micron droplet of cooking oil from a 2.5-micron particle of wildfire smoke — it reports mass concentration, not composition.

Carbon dioxide (CO₂) is a colorless, odorless gas produced by human metabolism and combustion. A person at rest exhales CO₂ at approximately 0.3–0.4 liters per minute, and the accumulation of exhaled CO₂ in an occupied room with limited ventilation is a direct proxy for the rebreathed fraction of air — the fraction of each inhalation that has recently occupied another person's lungs. The CO₂ concentration in outdoor ambient air is approximately 420 ppm; indoor concentrations in occupied spaces range from 500 ppm (well-ventilated) to over 3,000 ppm (poorly ventilated, densely occupied). ASHRAE Standard 62.1 recommends maintaining indoor CO₂ no more than 700 ppm above outdoor levels as an indicator of acceptable ventilation for occupant-perceived bioeffluent comfort. A CO₂ sensor reports a single gas concentration, not a particulate measurement; it requires a chemical detection mechanism that responds specifically to CO₂ and not to the dozens of other gases present in indoor air at comparable or higher concentrations.

Volatile organic compounds (VOCs) are a class of carbon-based chemicals that evaporate at room temperature from paints, adhesives, cleaning products, furniture, building materials, and personal care products. The term encompasses hundreds of individual compounds with widely varying toxicities — formaldehyde, benzene, toluene, xylene, acetone, ethanol, and terpenes are among the more common indoor species. A VOC sensor cannot identify which compounds are present or distinguish a hazardous compound (benzene at 50 ppb) from a nuisance compound (ethanol from hand sanitizer at 500 ppb) — it reports a single aggregated signal, typically expressed as a total volatile organic compound (TVOC) concentration in parts per billion or micrograms per cubic meter, that represents the sensor's electrical response to the mixture of all VOCs present, weighted by each compound's sensitivity factor.

PM2.5 and PM10: Laser Scattering Particle Counters

The particle sensor in a consumer air quality monitor is a miniature optical particle counter (OPC) based on laser light scattering. The sensor contains a laser diode — typically 650–680 nm (red) or 405 nm (violet) — a photodiode detector positioned at 90 degrees to the beam axis, and a small fan or resistive heater that draws an air sample through the sensing chamber at a controlled flow rate. When a particle passes through the laser beam, it scatters light in all directions, and a fraction of that scattered light strikes the photodiode, producing a voltage pulse whose amplitude is proportional to the particle's optical diameter. The sensor's microcontroller sorts each pulse into a size bin — typically PM1.0, PM2.5, PM10 — and increments the corresponding particle count. The particle count per unit volume, multiplied by an assumed particle density and shape factor, is converted to a mass concentration in µg/m³ using a calibration algorithm derived from measurements against a reference instrument in a controlled aerosol chamber.

Two sensor hardware platforms dominate the consumer market. The Plantower PMS5003 and its variants (PMS7003, PMS1003) use a 650 nm laser and a 0.1 L/min flow rate with a manufacturer-claimed minimum detectable particle size of 0.3 µm and a concentration range of 0–500 µg/m³. The PMS5003 is the sensor inside most consumer monitors under $200, including the PurpleAir and numerous budget monitors. The Sensirion SPS30 uses a 658 nm laser with proprietary contamination-resistance coatings on the optical surfaces — dust accumulation on the laser lens and photodiode is the dominant failure mechanism for OPCs in dusty environments — and a more sophisticated particle-sizing algorithm that Sensirion claims achieves ±10% mass concentration accuracy against a reference-grade instrument for PM2.5. The SPS30 appears in monitors from Airthings and several European manufacturers.

The critical limitation of consumer OPCs is that they measure optical particle size, not aerodynamic diameter, and they measure particle count, not mass. The conversion from count to mass assumes that the particles are spherical, have a density of approximately 1.0–1.65 g/cm³, and have a refractive index of approximately 1.5–1.6 — assumptions that are approximately true for ambient outdoor aerosol (sulfates, nitrates, organic carbon) but break down for specific indoor particle sources. Cooking aerosol from frying oil has a different refractive index and density than wildfire smoke, and neither matches the calibration aerosol. A consumer OPC placed next to a $20,000 reference-grade beta attenuation monitor (BAM) in an outdoor environment will typically track the BAM's readings with an R² of 0.85–0.95 but a slope that differs from unity by 20–40%, meaning the consumer sensor consistently reads high or low by a systematic factor that can be corrected with a post-processing calibration factor if a co-location reference measurement is available. In the absence of such correction, a reading of 25 µg/m³ from a consumer OPC should be interpreted as indicating that the PM2.5 concentration is somewhere between approximately 15 and 35 µg/m³ — a range that is sufficient to distinguish a clean room from a smoky one but insufficient for regulatory compliance determination.

CO₂: Non-Dispersive Infrared (NDIR) Sensors

The CO₂ sensor in a consumer air quality monitor is an NDIR (non-dispersive infrared) gas sensor — a technology that exploits the fact that CO₂ molecules absorb infrared radiation at a wavelength of approximately 4.26 µm, a wavelength at which water vapor, the most abundant infrared-absorbing gas in indoor air, has negligible absorption. The sensor consists of an infrared source — a miniature incandescent filament or an LED — that emits broadband IR radiation, an optical path through a sample chamber into which room air diffuses through a permeable membrane, an optical bandpass filter centered at 4.26 µm with a bandwidth of approximately 0.15 µm, and a thermopile or pyroelectric detector that measures the intensity of the transmitted radiation. The CO₂ concentration is calculated from the ratio of the transmitted intensity to the source intensity using the Beer-Lambert law: I = I₀ × e^(−α·c·L), where α is the absorption coefficient of CO₂ at 4.26 µm, c is the CO₂ concentration, and L is the optical path length.

The dominant NDIR sensor in consumer monitors is the Senseair S8 (and its miniaturized variants, the S8 LP and S8 0053), a Swedish-designed module with an optical path folded into a compact package roughly 34 × 21 × 14 mm. The S8 specifies an accuracy of ±40 ppm ±3% of reading over the 0–2,000 ppm range and ±5% of reading above 2,000 ppm, with a measurement interval of 2 seconds and a power consumption of approximately 18 mA at 3.3 V — low enough to operate for months on a small lithium battery. The Winsen MH-Z19B, a Chinese-manufactured alternative using similar NDIR principles, appears in lower-cost monitors and typically delivers accuracy of ±50 ppm ±5% of reading. Both sensors incorporate an automatic baseline calibration (ABC) algorithm that assumes the sensor will, at some point during a rolling 7–14 day window, be exposed to fresh outdoor air at approximately 420 ppm, and adjusts the zero point to match that assumed baseline. In a tightly sealed building where windows are not opened for weeks, the ABC algorithm will drift the baseline downward, producing CO₂ readings that are artificially low by 50–200 ppm — a systematic error that is invisible on the display but consequential if the monitor is used to assess ventilation adequacy.

NDIR sensor drift over time is dominated by two mechanisms: contamination of the optical surfaces (dust accumulation on the source window, detector window, or reflective walls of the optical cavity, which attenuates the beam and mimics CO₂ absorption) and aging of the infrared source (a tungsten filament whose emissivity declines as the filament evaporates, reducing the source intensity and mimicking increased CO₂). Both mechanisms produce a positive drift — the reported CO₂ increases over time as the true concentration remains constant. The ABC algorithm compensates for drift by periodically resetting the baseline to the assumed outdoor concentration, but the compensation is a linear offset correction, not a slope correction, and a sensor that has drifted by more than approximately 200 ppm from its factory calibration will typically require manual recalibration — exposing it to fresh outdoor air or a zero-CO₂ gas and initiating a calibration routine through the monitor's interface.

VOC: Metal Oxide Semiconductor (MOS) Sensors

The VOC sensor in a consumer monitor is a metal oxide semiconductor (MOS) gas sensor — a technology that originated in the 1960s for industrial combustible gas detection and was miniaturized in the 1990s for automotive cabin air quality control. The sensor consists of a heated metal oxide film — typically tin dioxide (SnO₂) doped with palladium or platinum catalysts — deposited on a micro-hotplate that maintains the sensing layer at 200–400°C. At this temperature, oxygen molecules from the air adsorb onto the metal oxide surface and capture electrons from the conduction band, creating a depletion layer that increases the electrical resistance of the film. When a VOC molecule contacts the heated surface, it reacts with the adsorbed oxygen, releasing the captured electrons back into the conduction band and decreasing the resistance. The magnitude of the resistance change is proportional to the concentration of reducing gases (VOCs are reducing agents in this chemical context) in contact with the sensor. The sensor's output — a raw resistance value — is converted to a TVOC concentration in ppb or µg/m³ using a calibration curve established by the sensor manufacturer against a reference gas mixture, typically a blend of several common VOCs at known concentrations.

The dominant MOS sensor in consumer monitors is the Sensirion SGP40 (or the VOC channel of the Sensirion SEN5x environmental sensor module, which integrates the particle, CO₂, and VOC sensors on a single board), with the ams CCS811 and Bosch BME688 appearing in lower-cost and older designs. The SGP40 specifies a response time of less than 10 seconds to a step change in VOC concentration and a long-term drift of less than 15% per year under typical indoor operating conditions — a specification that Sensirion achieves through a proprietary algorithm that compensates for the sensor's sensitivity degradation over time by tracking the sensor's response to a low-level hydrogen pulse generated internally by the micro-hotplate.

The fundamental limitation of MOS VOC sensors is not accuracy — under controlled laboratory conditions, a calibrated SGP40 can track a reference photoionization detector (PID) with reasonable fidelity — but sensitivity non-uniformity. The metal oxide surface responds to different VOC species with different sensitivities, and the sensitivity ratios can differ by an order of magnitude. The SGP40 is approximately 8–10 times more sensitive to ethanol than to toluene per unit concentration; a monitor that reports a TVOC of 500 ppb may be indicating 50 ppb of ethanol (from a surface wiped with an alcohol-based cleaner) or 500 ppb of toluene (from a freshly painted wall), and the user cannot determine which from the displayed number alone. The TVOC reading is a weighted sum of all VOCs present, and the weighting is determined by the electrochemical properties of the sensor — not by the toxicity of each compound. A compound with high toxicity but low MOS sensor sensitivity (formaldehyde is a notable example; MOS sensors have poor sensitivity to formaldehyde relative to ethanol and acetone) will be under-reported relative to its health significance. For this reason, an MOS-based TVOC reading should be interpreted as a relative indicator of changes in VOC concentration — a spike from 100 ppb to 500 ppb indicates that a VOC source has been introduced into the room; a drop from 500 ppb back to 100 ppb indicates that ventilation has diluted or removed the source — rather than as an absolute concentration measurement suitable for exposure assessment.

Accuracy Against Reference Instruments

The South Coast Air Quality Management District (AQMD) operates an Air Quality Sensor Performance Evaluation Center (AQ-SPEC) that has tested over 50 consumer air quality monitors against Federal Equivalent Method (FEM) and Federal Reference Method (FRM) instruments in both laboratory chamber and field co-location configurations. The published results establish performance tiers that are broadly consistent across multiple testing rounds:

PM2.5: The best-performing consumer monitors achieve R² values of 0.90–0.98 against FEM beta attenuation monitors in field co-locations lasting 4–8 weeks, with slopes of 0.8–1.2. The PurpleAir PA-II, after application of the EPA's correction factor (a nonlinear correction formula published by the EPA's Office of Research and Development in 2021 that adjusts for the sensor's known high bias at low humidity and low bias at high humidity), performs at the upper end of this range. Uncorrected consumer OPCs typically read 20–40% higher than FEM instruments at concentrations below 20 µg/m³ and converge to within ±10% at concentrations above 50 µg/m³. The systematic bias is sensor- and manufacturer-specific; applying a correction factor derived for one sensor model to data from a different model is not valid.

CO₂: NDIR sensors are the most accurate of the three consumer sensor types when properly calibrated. AQ-SPEC testing shows that consumer NDIR sensors achieve accuracy of ±50–75 ppm at 1,000 ppm against a reference nondispersive infrared analyzer, with R² values typically above 0.98. The dominant source of error is not the sensor hardware but the ABC algorithm — in a building where the sensor is never exposed to outdoor air, the baseline drifts downward and the reported CO₂ is low by 50–200 ppm. Manual recalibration by exposing the monitor to outdoor air for 30 minutes eliminates this error for a period of weeks.

VOC: MOS sensor performance against a reference PID is inconsistent and compound-dependent. A consumer monitor that reports 500 ppb TVOC in a room where a reference instrument measures 450 ppb of a toluene-ethanol mixture may report 200 ppb in a room where the reference instrument measures 450 ppb of a formaldehyde-acetone mixture — the sensor's cross-sensitivity to formaldehyde is low, and the TVOC number understates the true concentration. The TVOC reading is most useful as a trend indicator (rising or falling over hours or days) and least useful as an absolute value for comparison against health guidelines. The WHO and EPA do not publish indoor TVOC guidelines for MOS sensor readings because the sensor response does not correspond to the toxicologically relevant metric.

Sensor Placement

The placement of an air quality monitor determines whether the readings represent the air the occupants actually breathe or a localized anomaly near a pollutant source or a ventilation inlet. The monitor should be placed in the room where the occupants spend the most waking hours — typically the living room, family room, or home office — at a height of 3–5 feet, corresponding to the breathing zone of a seated adult. Placement should avoid the following locations, each of which produces readings that are not representative of bulk room air:

Within 3 feet of a kitchen. Cooking emits particles, CO₂ from gas combustion, and VOCs from heated oils and seasonings. A monitor placed within the convection plume of a stovetop will register transient PM2.5 spikes exceeding 500 µg/m³ during high-heat frying and CO₂ elevations of 200–500 ppm from a gas burner operating without a range hood. These readings accurately reflect the air in the kitchen at that moment but do not represent the air in the adjacent living space after mixing.

Within 5 feet of an exterior door or an operable window. Outdoor air infiltrating through a door or window carries ambient PM2.5 and CO₂ that differ from the indoor air by a variable amount depending on outdoor conditions. A monitor near a window that is occasionally opened will show abrupt concentration changes that are not representative of the indoor air elsewhere in the building.

Within 3 feet of an HVAC supply register. Conditioned air emerging from a supply register has been filtered by the HVAC system's air filter, which removes a fraction of the particulate matter — typically 20–60% for a MERV 8 filter, 60–90% for a MERV 13 filter — and contains outdoor air if the system includes an economizer or outdoor air intake. The air at the register is cleaner than the room-average air, and a monitor placed in the direct airflow will under-report the true PM2.5 concentration in the breathing zone.

On an exterior wall. Exterior walls are colder than interior walls in winter and warmer in summer. The thermal boundary layer at the wall surface affects the local concentration of VOCs (which desorb from building materials at higher temperatures) and can produce condensation on the sensor's optical surfaces if the wall temperature drops below the indoor dew point.

In direct sunlight. Solar heating of the monitor enclosure raises the internal temperature by 5–15°C above room ambient, which shifts the baseline of the MOS VOC sensor (MOS resistance is strongly temperature-dependent) and can produce a positive PM2.5 artifact if thermal convection inside the enclosure alters the flow rate through the particle sensor.

For a multi-room assessment, a single monitor can be moved between rooms and allowed to equilibrate for a minimum of 30 minutes in each location — the time required for the sensor's internal sample chamber to flush with the new room air and for the displayed reading to stabilize. Simultaneous monitoring of multiple rooms requires multiple monitors; a single monitor reporting an average of conditions in rooms that it is not in is reporting an extrapolation, not a measurement.

Interpreting the Numbers: What a Reading Means and What It Does Not

A consumer air quality monitor displays three numbers — PM2.5, CO₂, and TVOC — that appear to have the same epistemological status: each is a measurement of a pollutant concentration. They do not. The PM2.5 reading from a laser OPC is an estimate of mass concentration derived from an optical particle count, accurate to approximately ±30% for ambient aerosol and less accurate for specific indoor particle sources. The CO₂ reading from an NDIR sensor is a direct gas concentration measurement, accurate to approximately ±5% when calibrated. The TVOC reading from an MOS sensor is a weighted electrical response to a mixture of unknown composition, accurate as a trend indicator and unreliable as an absolute concentration.

The practical interpretation framework, consistent with the sensor technologies' capabilities and the published health guidelines, is:

PM2.5 below 12 µg/m³: Consistent with the EPA's annual National Ambient Air Quality Standard. Indoor concentrations consistently below this level indicate effective filtration and minimal indoor particle generation. PM2.5 12–35 µg/m³: Moderate elevation, consistent with the EPA's 24-hour standard range. Investigate potential indoor sources — candles, cooking with inadequate ventilation, vacuuming without a HEPA-filtered vacuum. PM2.5 above 35 µg/m³: Elevated. If sustained for more than one hour, consistent with a significant indoor source (frying, fireplace use, infiltration of wildfire smoke) or a sensor malfunction (dust accumulation on the laser optics). Cross-check against outdoor PM2.5 data from the nearest AirNow monitoring station before attributing an indoor PM2.5 spike to an indoor source; outdoor wildfire smoke can elevate indoor PM2.5 to 50–150 µg/m³ even in a closed building.

CO₂ below 800 ppm: Well-ventilated space with respect to occupant-generated bioeffluents. CO₂ 800–1,200 ppm: Marginally ventilated; approximately 1–2% of each inhalation has been previously exhaled. CO₂ above 1,200 ppm: Poorly ventilated; the rebreathed fraction exceeds 2% and cognitive performance decrements are detectable in controlled studies — a 2016 study by Allen et al. in Environmental Health Perspectives found that cognitive function scores on a strategic simulation were 50–60% lower at 1,400 ppm compared to 550 ppm, with the largest decrements in crisis response, information usage, and strategy. CO₂ above 2,500 ppm: The rebreathed fraction exceeds 4%, complaints of stuffiness and odor are nearly universal, and the space should not be occupied for extended periods without supplemental ventilation. These thresholds assume a monitor with a functioning ABC algorithm and recent exposure to outdoor air; if the monitor has been in continuous indoor use for months without recalibration, the reported CO₂ may be 50–200 ppm low, and the true concentration may be correspondingly higher.

TVOC: The absolute number is less informative than the trend. A stable TVOC reading between 100–500 ppb that does not change over days or weeks indicates a steady-state balance between indoor VOC emission sources (furniture, building materials, occupants) and ventilation removal. A transient spike — 500 to 2,000 ppb over 10–30 minutes — indicates an acute emission event: cleaning with solvent-based products, painting, cooking with oils at high temperature, or the introduction of a new VOC-emitting product into the room. A sustained elevation that persists for days indicates a continuous source — off-gassing from new furniture or cabinetry, a solvent spill that has not been cleaned, or a ventilation system that is not exchanging air. In all cases, the appropriate response to an elevated TVOC reading is ventilation: opening windows and operating exhaust fans. The TVOC number alone, without compound identification, does not support a more specific remediation action.

Product Categories: Standalone Monitors and Multi-Sensor Integration

Consumer air quality monitors fall into three broad categories distinguished by sensor quality and feature set. Entry-level monitors ($50–120) typically use a Plantower PMS5003-class particle sensor, a Winsen MH-Z19B or equivalent NDIR CO₂ sensor, and a basic MOS VOC sensor (often an older-generation CCS811 or a discrete tin-dioxide sensor without temperature compensation). These monitors provide trend-level data for all three pollutants and are sufficient for identifying large changes in indoor air quality — a PM2.5 spike during cooking, a CO₂ buildup in a closed bedroom overnight — but the TVOC readings below approximately 200 ppb may be indistinguishable from sensor noise, and the CO₂ readings may be unreliable without periodic outdoor-air calibration.

Mid-range monitors ($150–300) use Sensirion SPS30 or equivalent particle sensors, Senseair S8 NDIR CO₂ sensors, and Sensirion SGP40 MOS VOC sensors with factory-applied calibration. The Airthings View Plus adds a radon sensor (pulsed ion chamber) to the standard three-sensor suite, making it a four-pollutant monitor that combines the functions of a radon detector and an air quality monitor in a single instrument — an appealing integration for homeowners who would otherwise purchase both devices separately. The IQAir AirVisual Pro uses a proprietary laser particle sensor with a larger sensing chamber and a higher flow rate than the Plantower modules, achieving better low-concentration sensitivity (detection limit approximately 1 µg/m³) and better correlation with FEM instruments (R² typically above 0.92 in AQ-SPEC field tests) than most sub-$200 monitors. Both the Airthings and IQAir monitors provide Wi-Fi connectivity, smartphone apps with historical data graphing, and integration with IFTTT and home automation platforms for triggering ventilation or filtration equipment based on sensor readings.

Research-grade consumer monitors ($300–500+) bridge the gap between consumer instrumentation and reference-grade measurement. The Atmotube PRO is a portable, battery-powered monitor using Sensirion sensors with individual factory calibration certificates; the PurpleAir PA-II-Flex uses dual Plantower PMS5003 sensors in a redundant configuration that allows the user to detect sensor failure when the two sensors diverge. These monitors are suitable for citizen science projects, indoor air quality investigations where sensor accuracy must be documented, and post-remediation verification of air sealing or filtration upgrades.

Sensor Lifespan, Maintenance, and When to Replace

The laser particle sensor has a finite operating life determined primarily by the degradation of the laser diode and the accumulation of particulate contamination on the optical surfaces. The Plantower PMS5003 is rated for approximately 3 years of continuous operation at typical indoor particle concentrations; at elevated concentrations (chronic PM2.5 above 50 µg/m³, as may occur during a prolonged wildfire season), the rate of optical contamination accelerates, and the sensor may require replacement after 1–2 years. The failure mode is a gradual positive drift — the sensor reports increasing PM2.5 concentrations as light scattered from contamination on the optics is misinterpreted as particle scattering — which is difficult to distinguish from an actual increase in airborne particulate matter without a co-located reference measurement. A sensor that reads 15–25 µg/m³ in a room that has been unoccupied for 24 hours with windows closed and HVAC fan running continuously (conditions under which PM2.5 should be below 5 µg/m³) is likely exhibiting optical contamination drift and should be replaced or cleaned if the manufacturer provides a cleaning procedure.

The NDIR CO₂ sensor has a longer operational life — Senseair specifies 15+ years for the S8 under normal indoor conditions — but the ABC algorithm's assumptions must be maintained for the sensor to remain accurate. A monitor that is deployed in a building that is continuously occupied and never ventilated with outdoor air (a data center, a windowless conference room, a basement apartment with recirculating HVAC) will experience ABC algorithm drift and must be manually recalibrated monthly or quarterly by exposing the monitor to outdoor air for 30 minutes.

The MOS VOC sensor degrades through two mechanisms: poisoning of the catalytic metal oxide surface by siloxanes (from personal care products, deodorants, and silicone-based lubricants) and halogenated compounds, and gradual drift of the micro-hotplate temperature control circuit. Sensirion estimates the SGP40's operational life at 10+ years under typical indoor conditions, with the internal compensation algorithm maintaining accuracy within ±15% of the factory calibration over that period. However, a sensor that has been exposed to a high-concentration VOC event — a paint spill, a solvent cleaning session in an unventilated room — may exhibit a temporary shift in baseline that requires 24–48 hours of operation in clean air to recover, and a sensor exposed to siloxane vapors at elevated concentrations (from hair spray or silicone caulk curing in a closed room) may experience permanent sensitivity degradation that is not recoverable.