DTT is a reaction measured under conditions, not an illness meter
How to translate aerosol DTT oxidative potential begins with a useful distinction: a laboratory reaction is not the same thing as an observed illness. Dithiothreitol depletion is used to examine how redox-active material extracted from particles consumes a reagent under controlled conditions. A scientifically accurate account must identify the assay, the corrected reaction rate, the air sample represented, and the difference between per-air-volume and per-particle-mass normalization. Without those elements, two reports can display large numbers that answer different questions.
The most common reading failure is denominator confusion. A reported reaction of one nanomole per minute describes what happens in a reaction vessel. Dividing an appropriately scaled whole-sample result by the amount of air collected creates an air-volume-normalized activity, commonly expressed as nmol min−1 m−3. Dividing by the mass of sampled particulate matter instead produces an intrinsic or mass-normalized activity, commonly nmol min−1 µg−1. Neither should be labelled simply ‘toxicity’. They help answer different environmental questions and remain specific to the conditions of the DTT assay.
Begin with a teaching example. An experiment estimates sample DTT depletion as 1.35 nmol/min and a reagent blank as 0.15 nmol/min, giving 1.20 nmol/min after blank correction for the assayed, properly scaled filter equivalent. If that entire filter represented 24 cubic metres of ambient air, the air-volume-normalized activity is 1.20 divided by 24, or 0.050 nmol min−1 m−3. If the filter carried 240 micrograms of PM, its mass-normalized activity is 1.20 divided by 240, or 0.0050 nmol min−1 µg−1. Every number is a teaching value; no ambient monitoring observation is being asserted.
That calculation already raises essential laboratory questions. Was the whole extract tested, or only one aliquot? Did the technician apply the dilution factor? Was the blank from reagents or a filter collected and handled in the field? Which part of the reaction curve was used to fit a depletion rate? Were particles extracted in water, a mixed solvent, or tested in suspension? Can a response be compared with an assay from another laboratory? We work from the first principles of the reaction through defensible interpretation so students, researchers and editors can translate unfamiliar reports without exaggeration.
1. Oxidative potential is an assay result
The phrase aerosol oxidative potential describes a particle sample’s capacity to drive reactions under a specified laboratory assay, not a complete measurement of biological toxicity. The dithiothreitol assay quantifies depletion of a reducing reagent as the sample reacts in a controlled solution. It yields an operational response governed by assay chemistry.
Do not translate a high value as a directly measured cellular injury rate or a diagnosis of disease. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
2. DTT is a reagent, not a pollutant
Dithiothreitol, abbreviated DTT, is deliberately added to the laboratory reaction; the sampler does not collect DTT from urban air. The sample’s redox-active constituents accelerate loss of DTT under the specified conditions, and the analytical procedure measures that loss.
An article that calls DTT an airborne contaminant has reversed the experiment’s direction. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
3. What depletion means
If an aliquot begins with 100 nmol of DTT and has 78 nmol at the chosen later time, the measured loss is 22 nmol before blank and time normalization. The reading must be interpreted against the reaction time and the corresponding reagent-only control.
A 22 nmol decrease over 10 minutes is not the same depletion rate as a 22 nmol decrease over 40 minutes. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
4. Calculate a rate
A DTT decline from 90 to 66 nmol over 20 minutes corresponds to 24 divided by 20, or 1.2 nmol per minute, on a straight-line assumption. If the time course curves sharply, the average two-point slope may not represent a stable rate; evaluate the working range.
The unit nmol/min is an absolute rate for the material actually in that reaction vessel. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
5. Blank subtraction
A reagent blank losing 0.15 nmol per minute and sample reaction losing 1.35 nmol per minute give a blank-corrected activity of 1.20 nmol per minute. Record whether the instrument or spreadsheet already subtracts the blank before applying a further correction.
Subtracting the same blank twice will create artificially small activity. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
6. Aqueous extraction
A water extract preferentially represents water-soluble or water-extractable redox-active material under the procedure’s conditions. Particle components remaining on the filter or inaccessible in the solution may not participate in this measurement.
Name the result water-soluble DTT activity rather than whole-particle activity if only the filtrate was tested. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
7. Total-particle approaches
Protocols that retain filter material or use other extraction conditions can access additional less-soluble reactive fractions. Comparing an aqueous extract method with a total-particle method as though both measured the identical chemical universe is misleading.
State solvent, whether solids were removed, reaction pH and recovery limits. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
8. Filter punch representation
A circular filter punch may represent only a known fraction of the full collected area and therefore only a fraction of sampled particle mass. For a uniform deposit, a punch representing one quarter of the filter area contains approximately one quarter of the deposited mass; nonuniform loading weakens this assumption.
Account for face velocity, deposit geometry and actual punch area rather than treating area fraction as infallible. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
9. Extract volume accounting
If particles from a full filter are extracted into 10 mL and a 1 mL aliquot is analyzed, the measured reaction reflects one tenth of the extract under homogeneous conditions. Convert the aliquot activity to a full-filter equivalent using the documented dilution and aliquot factors when scientifically appropriate.
An unrecorded aliquot factor can create a tenfold error without any instrument malfunction. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
10. Air-volume normalization
A blank-corrected activity representing an entire filter of 1.2 nmol/min collected from 24 m³ of air corresponds to 0.050 nmol/min/m³. This air-volume-normalized activity describes the reactive response per volume of air sampled, after all extraction scaling has been performed.
Do not divide by air volume before restoring from a small assay aliquot to the sampled fraction. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
11. Mass normalization
If the full sampled filter carries 240 µg of particles and has 1.2 nmol/min activity, mass-normalized activity is 0.0050 nmol/min/µg. That number characterizes assay response per collected PM mass, not overall pollutant concentration in ambient air.
The mass-normalized ranking may differ substantially from the per-air-volume ranking. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
12. A comparison with equal mass
Suppose site A has 3 nmol/min from 300 µg particle mass and site B has 2 nmol/min from 100 µg. A gives 0.010 nmol/min/µg while B gives 0.020; the smaller total assay signal can have greater intrinsic activity per mass.
Always calculate denominators rather than equating the largest total reaction signal with the most reactive particles. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
13. A comparison with equal air volume
Two filters each sampled 20 m³ of air; activities are 0.6 and 1.5 nmol/min after all corrections. The per-volume activities are 0.030 and 0.075 nmol/min/m³, respectively, so the second air sample has the larger method-specific response per ambient volume.
An equal sampling volume simplifies comparison but does not remove differences in extraction recovery. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
14. Dilution into assay range
A concentrated extract may deplete the reagent too quickly or change the reaction kinetics; dilution can help keep observations in the validated linear working range. Record each dilution factor and apply it exactly once to calculate a comparable whole-sample activity.
If the response per unit input mass changes systematically after dilution, investigate matrix effects or nonlinear chemistry. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
15. Multiple reaction timepoints
An assay measured at 0, 5, 10 and 15 minutes may provide a more defensible rate than relying on two arbitrary endpoints. Fit a slope only across the validated time region and inspect residuals; a high coefficient of determination does not make a chemically invalid range valid.
Report the selected time interval rather than hiding it behind a single slope. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
16. DTNB detection chemistry
Some procedures use DTNB and optical absorbance to quantify residual DTT indirectly through a colour-producing reaction. Translate the absorbance reading with the method’s calibration before reporting moles remaining; the optical signal is not itself a unit of oxidative potential.
Wavelength, calibration and reagent chemistry belong in the analytical method, not silently in the final number. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
17. Optical blanks and colour
Dark-coloured extracts, turbidity or chemically interfering substances can affect optical measurement and alter apparent residual DTT. Use appropriate sample and optical blanks under the procedure rather than assuming every absorbance change represents reaction of DTT.
This is an analytical specificity question, separate from real redox activity. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
18. Calibration range
A standard curve relates instrument response to a known amount or concentration of the measured reaction product. Keep readings inside the validated range, record regression fit and blanks, and propagate uncertainty when converting optical units to nmol.
Extrapolating far beyond the standards can turn a precise-looking number into an unsupported estimate. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
19. Replicate reactions
Two or three technical replicates test variation in laboratory preparation and reaction measurement for a particular extract. They do not by themselves measure spatial variation in the atmosphere or day-to-day variability at a location.
Distinguish technical repeatability from variation among independent ambient samples. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
20. Positive control
A stable positive control helps detect assay batches that respond too weakly or strongly to a known redox-active standard. A passing positive control supports procedure consistency but does not guarantee recovery for every compositionally complex ambient extract.
Report control acceptance windows and investigate excursions before interpreting subtle site differences. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
21. Reagent blank
A blank reaction without particle extract estimates background depletion caused by assay reagents and conditions. Because blank reaction may vary with reagent age, pH or temperature, maintain appropriate batch-matched controls.
A historical blank should not automatically replace a failed concurrent blank. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
22. Field filter blank
A field blank follows collection handling without normal air sampling, helping reveal contamination introduced during transport, storage or handling. Its preparation may differ from a pure reagent blank, so the two should not be silently interchanged.
If field blanks show appreciable activity, include the resulting correction and uncertainty or explain why results are qualified. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
23. Storage effects
Filter extracts and particles may change chemically in storage through oxidation, loss of volatile components or container interactions. A sample analyzed the next day may not be perfectly comparable to one stored for months if stability has not been demonstrated.
Report storage temperature, duration, container and any tested holding-time limits. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
24. Temperature control
Reaction kinetics typically depend on temperature; temperature shifts can affect the measured depletion slope even if the aerosol composition is unchanged. Maintain and document the assay temperature and do not treat a value from an uncalibrated warmer batch as chemically identical.
A reanalysis after restoring a common temperature may separate assay drift from a genuine environmental pattern. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
25. pH and buffer
The protonation state and redox chemistry of reagents and particle constituents depend on the solution environment. Changing buffer concentration or pH can alter what the DTT assay reports, even when the starting aerosol is chemically identical.
Describe assay conditions explicitly before comparing methods from different laboratories. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
26. Reaction oxygen
Oxygen and electron-transfer pathways can participate in redox cycling of sample constituents. A reagent reaction conducted under materially different aeration conditions may produce different depletion kinetics.
Use documented mixing and vessel conditions rather than presuming oxygen exposure has no effect. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
27. Metal sensitivity
Soluble transition metals such as copper and manganese can contribute to DTT consumption in some particle mixtures. Their contribution depends on concentration, chemical complexation and solution conditions, not on a universal fixed fraction.
A high metal concentration is a clue about potential reactivity but not a complete substitute for assay measurement. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
28. Organic redox compounds
Certain quinones and other organics can drive redox cycling and affect a sample’s DTT response. A carbon-rich extract need not be uniformly reactive; molecular identity and chemical aging matter.
Do not use total organic carbon alone to infer the assay response without evaluating relevant chemistry. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
29. Metal chelation experiment
A controlled chelation test can probe how removing some metal availability changes DTT activity. The chelator itself may alter reaction conditions or interact with other constituents, so the decrease is not a perfect partition of source contributions.
Treat it as mechanistic evidence under specified conditions rather than an exact measurement of all metal-driven activity. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
30. Solubility and particle matrix
Two filters with similar metal loads may produce different assay activity because one releases metals more readily into water. An apparent chemical source difference can partly reflect extraction and complexation rather than atmospheric concentration alone.
Interpret dissolved and total elemental results separately, especially when linking to ICP-MS studies. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
31. Fine versus ultrafine
A submicron or ultrafine fraction can differ in chemistry and per-mass reactivity from a broader PM2.5 sample. Size classification and collection efficiency must be specified, as a particle number measurement cannot replace collected reaction-active mass.
Avoid saying ‘smaller always means more oxidative’ without actual size-resolved evidence. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
32. Comparing assay families
DTT, ascorbic acid depletion and cellular reactive oxygen species tests are different biological or chemical probes. They can rank samples differently because each is sensitive to different constituents and reaction pathways.
A value from one assay cannot be numerically converted to another without method-specific validation. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
33. Detection limit
A method detection limit describes the smallest response that can be distinguished under a specified decision rule; it does not guarantee accuracy near that boundary. A near-limit activity should be reported with uncertainty and the correct censoring notation, not rounded into a confident positive signal.
Do not turn a result below quantification into a scientifically meaningful zero. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
34. Uncertainty of a difference
When a sample blank-corrected activity equals a sample slope minus a blank slope, uncertainty in both matters. If independent slope uncertainties are 0.10 and 0.08 nmol/min, root-sum-square combined uncertainty is approximately 0.128 nmol/min.
Correlated errors require different treatment, so state whether independence is reasonable. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
35. Volume measurement uncertainty
A sampler reporting 24 m³ of air has calibration, flow and elapsed-time uncertainties. A 5 percent air-volume error approximately propagates into a 5 percent error in air-volume-normalized activity, with the inverse direction.
State whether reported flow was corrected to standard or ambient conditions; do not silently mix conventions. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
36. Normalization and causal interpretation
An air-volume-normalized DTT value rises when either more particle mass is collected per cubic metre or the collected material is more reactive per mass. The mass-normalized response helps distinguish these possibilities but adds its own denominator uncertainty.
Provide both metrics when the research question concerns concentration versus chemistry. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
37. Diurnal patterns
Short samplers may capture traffic peaks or sunlight-driven chemical aging, while a twenty-four-hour sample combines multiple regimes. A day-night comparison requires equivalent collection methods, extraction and handling to avoid confounding sampling conditions with atmosphere.
A time series alone does not identify which reaction pathways caused its pattern. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
38. Seasonal interpretation
Winter heating and summer photochemistry can change aerosol composition; temperature can also change sampling and chemistry. A high summer DTT signal cannot be assigned to secondary organic aerosol without source or chemical evidence.
Compare composition and meteorology before naming a dominant cause. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
39. Roadside versus background
A higher roadside value per cubic metre could reflect particle loading, particular reactivities or both. Compare mass-normalized activity and measured metals or organics; note site height, wind direction and distance from sources.
Do not equate geographic proximity with an identified biochemical mechanism. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
40. Correlations with PM2.5
A strong correlation between DTT per air volume and PM mass may reflect shared variation in loading rather than shared intrinsic reactivity. Examine mass-normalized DTT activity and composition alongside the per-volume value.
Correlation is descriptive evidence, not proof that a specific component mediates health outcomes. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
41. Public health communication
Oxidative potential is investigated as a possible dimension of harmful particle mixtures beyond their gravimetric mass. The DTT test is an acellular assay and cannot alone provide a population-specific disease risk coefficient.
Avoid giving advice that implies a laboratory reaction rate translates directly into clinical outcomes. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
42. Source attribution from DTT
If copper, organic markers and DTT activity rise together, several pathways may be plausible. A co-variation plot cannot separate metal causation from a common source carrying both metals and organic substances.
Mechanistic experiments, cross-validation and defensible receptor-model evidence are needed for stronger claims. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
43. Reviewing a sample spreadsheet
A reproducible workbook records sampling air volume, PM mass, filter fraction, extraction volume, aliquot volume, timepoints, calibration, blank slope and final units. Trace each value from raw measurements to corrected activity, then to per-volume and per-mass normalization.
If any intermediate scale is absent, request the method details rather than guessing the missing factor. Translate this result at three levels: what the laboratory read, what mathematical correction was performed, and what environmental statement that corrected result supports. Keeping these clauses separate prevents operational assay response from being renamed as a biological outcome.
44. Comparing published studies
A paper may report nmol/min/m³ while another gives nmol/min/µg; they are distinct quantities. Conversion requires matching PM mass concentration and method definitions, not merely moving a decimal point.
Check whether both studies analyzed water-soluble extracts and shared comparable reaction conditions. Where methods diverge, do not force false equivalence. Document the actual filters, extractions, reaction conditions and units; if those descriptions are unavailable, restrict the inference to the individual study. That limitation is more valuable than an unsupported cross-study ranking.
45. Worked interpretation of an increase
A city’s mean DTT per volume rises by 25 percent over a period while PM2.5 mass rises by 50 percent. Under stable denominators and assays, per-mass DTT activity could decline in relative terms; it does not necessarily mean chemistry became intrinsically more reactive.
Calculate both series using the same sampled periods before narrating a trend. Before presenting this in a comparison table, preserve the exact order of operations and the reported unit. An apparently small scaling choice can be larger than a real environmental difference. The reader should be able to reconstruct the result with a calculator using only the stated values and conditions.
46. What a defensible conclusion says
A method-specific result can quantify redox reaction in a defined extract, while chemistry and source analysis provide hypotheses about why it occurred. The strongest report separates measured reaction response from explanatory association and from downstream health inference.
State units, blanks, reaction conditions, extraction fraction, uncertainty and alternative mechanisms in the final summary. A useful second check is to separate analytical precision from environmental interpretation. Repeatability may be good even when the selected method excludes some reactive chemicals or responds differently to mixture chemistry. Confidence in a repeatable signal is not the same as confidence in its cause.
Trace the calculation backwards and solve four new problems
To verify a published oxidative-potential number, work backwards. Identify what the reaction instrument measured, the amount of DTT remaining at defined times, the calibration used to translate instrument signal into amount, the linear time interval, and the blank correction. Then restore the fraction of the sampled filter represented by the reaction vessel. Finally divide by either sampled air volume or particle mass, depending on the stated reporting target. Add uncertainty from blanks, sample collection, aliquot scaling and reaction slope; document any untested correlation between terms.
Independent test one: a full-filter equivalent produces 2.4 nmol/min after correction and represents 60 m³ of air. Its air-volume-normalized activity is 0.040 nmol min−1 m−3. If that filter holds 300 µg of collected particulate matter, its mass-normalized activity is 0.0080 nmol min−1 µg−1. The two results are not competing estimates; they have different denominators. Independent test two: a reaction sample loses DTT at 1.10 nmol/min and the reagent blank loses 0.22 nmol/min. The corrected value is 0.88 nmol/min, assuming the protocol licenses that subtraction.
Independent test three: after dividing by particle mass, site A has activity 0.012 nmol/min/µg and site B 0.006, but B has four times as much particulate matter per cubic metre. With comparable methods, B has twice the per-air-volume activity of A: the mass multiplication overcomes the lower intrinsic rate. Independent test four: two samples have equal DTT reaction rates, but one is an aqueous soluble fraction and the other a total-particle suspension. Can they be called chemically equivalent? No; assay accessibility and matrix definitions differ.
The portable reading strategy is the same one that works for scientific vocabulary and technical translation elsewhere: identify the substance, method, numerator, denominator, correction, uncertainty and boundary of inference. A number can be accurate for its protocol while becoming misleading in a sentence that silently changes what it is supposed to mean. The goal is not to make DTT sound complicated. It is to make the result auditable and useful without pretending it is a complete human health test.
Primary methods and useful connections
Methodological reading includes Fang and colleagues in Atmospheric Measurement Techniques (2015), Gao and colleagues in Atmospheric Measurement Techniques (2017), and the EPA research record for a source-oriented DTT study. For measurement context, read eduKate guides on trace metals, filter mass and sampled air volume, and PMF source factors. All worked numbers above are illustrations, not laboratory data.