Photoluminescence Quantum Yield: Read PLQY and Sphere Data

How do you read a photoluminescence quantum yield (PLQY) result? Start with the fraction of absorbed excitation photons that lead to emitted photons, then check how the experiment counted each side of that ratio. A sample that produces a bright photoluminescence spectrum has not necessarily converted light efficiently. A calibrated integrating-sphere measurement asks a much more specific question: of the light actually absorbed, what fraction reappears as measured luminescence under the stated conditions?

Readers searching for absolute PLQY, integrating-sphere quantum yield, fluorescence quantum yield calculations, absorbed-photon corrections, or photoluminescence measurement errors usually need a way to tell the physical result from instrument response. This guide shows how to read the excitation deficit, the emission spectrum, the blank, and the uncertainty before comparing materials. It treats numerical examples as illustrations rather than reports of measurements.

The aim is to translate technical charts, experimental notes, and scientific English into a defensible explanation that can be carried between disciplines and languages. We distinguish photon count from optical power, absolute yield from relative brightness, and a plausible interpretation from a proven physical mechanism. The same habits apply to dyes, phosphors, semiconductors, quantum dots, films, solutions, and other luminescent systems.

The central idea: count photons, not adjectives

Photoluminescence quantum yield is commonly written ΦPL = number of emitted photons divided by number of absorbed excitation photons. It is a dimensionless ratio, often displayed as a percentage. If a measurement supports 0.72, the direct interpretation is that about seventy-two photons were emitted for every hundred excitation photons absorbed in the conditions of that experiment. It does not mean seventy-two percent of the original optical energy was recovered, and it does not guarantee the same yield under another excitation intensity or in another solvent.

A photon at a longer emission wavelength carries less energy than a photon at a shorter excitation wavelength. Energy efficiency and photon efficiency are therefore different descriptions. This distinction matters particularly when a result table reports spectral power rather than photon counts. An honest explanation keeps the numerator and denominator visible, names the experimental conditions, and refuses to replace the actual definition with ‘how much the sample glows.’

Why brightness is not the same measurement

Imagine two films photographed under one ultraviolet lamp. The first looks bright from the camera position; the second looks dim. That appearance may reflect how much light the films absorb, their surface texture, angular emission, camera exposure, or the detector’s colour sensitivity. A high-yield sample that barely absorbs the lamp wavelength can appear less bright than a low-yield sample that absorbs nearly every incident photon. The visual ranking is not the quantum-yield ranking until absorption and detection have been accounted for.

Relative emission intensity is still useful when geometry, collection time, concentration, excitation and detection settings are controlled. It can show that a specimen changed after heating or aging. But even a reproducible intensity increase does not provide an absolute PLQY value by itself. The first habit of reading any chart is to ask whether its vertical axis represents calibrated photons, counts, watts, a detector voltage, or arbitrary units.

Start with the complete spectral story

A PLQY study involves an excitation source, light transmitted or reflected without being absorbed, and light newly emitted by the sample. The relevant spectra can overlap. A short-wavelength excitation peak can sit near a longer-wavelength luminescence band; stray source light may enter the emission integration window. Spectral separation and instrument correction determine whether the apparent emitted area is truly luminescence. Never integrate every visible curve into one grand photon total.

Read the legend before inspecting the tallest feature. Determine which trace represents an empty sphere or blank, which represents the sample, and whether the plotted intensity has been corrected for detector response. Identify the units and spectral region of the horizontal axis, and record the excitation wavelength and its bandwidth. If these are missing, a precise efficiency percentage deserves a cautious rather than celebratory caption.

Convert radiant power into photon numbers correctly

Many detectors measure a quantity proportional to optical power at each wavelength. The energy of one photon is hc/λ, where h is Planck’s constant, c is light speed, and λ is the wavelength. A spectral power distribution must therefore be weighted appropriately to convert it into photon rate before comparing absorbed and emitted photon totals. Dividing a total emission power by a total excitation power estimates an energy ratio, not automatically the quantum yield.

Pay attention to the axis definition before applying any conversion. A spectrum expressed per nanometre cannot simply be integrated with the same numerical operation as a spectrum expressed per electronvolt without a Jacobian transformation. Spectrometer software may already report photon-corrected intensity; repeating the correction would distort the answer. A methods statement should identify exactly what calibration and integration were performed, so another researcher can audit the number.

What an integrating sphere changes

An integrating sphere has a reflective interior that redistributes light through many reflections before detection. The design reduces dependence on the direction in which light leaves the specimen, allowing a more complete assessment than a single small collection angle. Yet ‘sphere’ is not a synonym for ‘perfect capture.’ Ports, baffles, detector positions, coating reflectance, the sample holder, and wall contamination affect the response. A sphere measurement still needs calibration and a carefully matched blank.

Two sphere instruments may differ in geometry or in how they calculate the result. One places a sample in the direct excitation path; another measures indirectly after the light has first struck the wall. A third uses multiple configurations. Before comparing published quantum yields, ask which setup was used and whether the reported corrections match it. An instrument’s output number becomes credible through a described acquisition procedure, not through the shape of the sphere alone.

Read the blank as an experiment, not as decoration

A blank can be an empty holder, an identical cuvette filled only with solvent, an unsubstituted substrate, or another optical match to the specimen. It estimates what the excitation beam and surrounding apparatus contribute without the luminescent material. If the sample and blank scatter or absorb light differently, subtraction can produce a biased absorbed-photon number. A transparent blank does not automatically represent a strongly scattering ceramic powder, for example.

The blank spectrum should be acquired with stable illumination and appropriate settings, ideally close in time to the sample measurement. Look for changes in source intensity, sphere background, and reflected excitation light. A quantum yield that appears unusually high because the blank was mismatched is not rescued by quoting three decimal places. Ask whether the blank is physically comparable before trusting the resulting arithmetic.

How the absorbed-photon denominator is formed

One common absolute-measurement logic compares the excitation light remaining with the sample present against the excitation light observed in a corresponding reference or blank configuration. The difference estimates photons absorbed by the specimen, after corrections specific to the sphere method. If the sample absorbs only a tiny fraction, that denominator is the difference between two large signals; it can be extremely sensitive to drift, small calibration errors, and scattering changes.

A useful diagnostic is to show the source and residual excitation areas, not only their difference. Suppose the reference produces an illustrative 1,000,000 corrected photon units and the sample configuration produces 990,000. The nominal absorption is 10,000 units, only one percent of the incoming reference. A small one-percent shift in instrument response could therefore be comparable to the entire absorption signal. A confident reported yield would need especially convincing controls.

How the emitted-photon numerator is formed

The luminescence numerator comes from spectral emission attributable to the sample after subtraction of stray light and instrument background. It must be integrated over enough wavelength range to include the relevant emission, including red or near-infrared tails if those are part of the signal. A narrow scan can omit real photons and underestimate yield. Extending the scan indiscriminately into an uncalibrated, noisy region can create the opposite problem if spurious background is counted.

If multiple emitting species overlap, total PLQY can still be defined for the measured system, but assigning a separate yield to one component requires spectral separation and assumptions. Emission shoulders are not necessarily separate molecules. Compare excitation dependence, controlled mixtures and time-resolved evidence before naming species. A robust reading separates the defensible integrated photon result from a tentative chemical explanation for why the spectrum has its shape.

Direct versus indirect excitation

Some protocols distinguish a configuration in which the excitation beam hits the sample directly from one in which it illuminates the sphere wall first. The difference is important because sample absorptance can depend on angle, path length and illumination distribution. A material that is optically transparent in a thin layer may behave differently as an opaque powder or textured film. A protocol designed for a solution cannot automatically be transferred to a porous scattering material.

A schematic or methods section should state sample position, illumination direction, port configuration, and how the blank was placed. If direct and indirect readings disagree, investigate geometric effects rather than averaging blindly. The discrepancy may tell you something useful about the sample, but it does not by itself prove an exotic emissive process. Treat configuration as an experimental variable that needs to be reported.

Account for wavelength-dependent detector sensitivity

A silicon detector and a near-infrared detector do not respond equally at all wavelengths, and optical gratings, filters and fibre couplings introduce their own spectral bias. An uncorrected spectrum can make one emission colour look dominant because the instrument is particularly sensitive there. Absolute quantum-yield calculations need an appropriate spectral-responsivity correction across the observed bands. The correct calibration depends on the measurement chain, not just the photoluminescent material.

Inspect whether the correction is valid at both excitation and emission wavelengths. Calibration outside the detector’s reliable range is not solved by extending a fitted curve indefinitely. If the instrument switches detectors or gratings part-way across the spectrum, a discontinuity can appear in the signal. State that limitation explicitly if it affects the integrated area. The goal is a meaningful uncertainty interval, not a deceptively smooth curve.

Self-absorption and re-emission complicate the numerator

A photon created inside a concentrated dye, thick film or powder may be reabsorbed before it escapes to the sphere detector. It may disappear through a nonradiative path or be emitted again at another wavelength. The collected spectrum can therefore be red-shifted and reduced relative to the material’s first emission event. This process is often called reabsorption or self-absorption, and it becomes important when absorption and emission bands overlap.

A correction requires assumptions or additional measurements. It is not valid to declare that every red-shifted band proves a new emitter, or to apply a universal percentage correction to every solid. Examine specimen thickness, concentration, emission geometry and published correction method. A paper that reports both the observed yield and a model-corrected estimate may be more informative than one that reports only the preferred larger number.

Scattering samples deserve special caution

Particles and rough films can redirect excitation light into paths that differ from the transparent-blank measurement. Their emitted photons may also be redistributed differently by the sphere. The resulting error can act on both numerator and denominator, so repeating an arithmetic calculation with more decimal places does not solve it. The relevant controls are optical: matching scattering behaviour where possible, documenting sample location and testing geometry dependence.

Researchers sometimes position a powder at the sphere centre, mount it at a port, or place it on a holder within the illuminated region. Those choices can change how the blank should be designed. A strong report describes the physical configuration well enough for another laboratory to approximate it. Without such information, compare only broad patterns between materials and reserve exact rankings for genuinely comparable setups.

Low absorption can make a plausible number fragile

Suppose a faintly absorbing film produces a nominal 65% PLQY. The value may be correct, but look at the raw absorbance or excitation deficit before trusting it. When absorption is near the noise floor, very small drift in lamp intensity or blank placement produces a large relative change in the denominator. High emission signal-to-noise ratio does not fix low confidence in absorbed photons. The two sides of the calculation must be examined separately.

Repeat measurements at sensible absorption levels where sample chemistry permits, and inspect whether the inferred yield stabilises. Diluting too far can improve optical linearity but worsen denominator uncertainty; concentrating too far can introduce aggregation and reabsorption. The useful concentration lies in a measurement window supported by both sample behaviour and instrumental precision. An uncertainty budget should expose the trade-off.

High optical density introduces a different problem

At very high absorption, nearly all excitation photons may be removed near the entrance surface. The experiment can become sensitive to illumination position, inner-filter effects, local heating and nonuniform photochemistry. A thick specimen may appear to have a different PLQY simply because photons are created and trapped in a different region. A high absorption fraction does not automatically make the result more reliable.

Interpret concentration series together with emission shifts, spectral width and stability. If the spectrum changes with concentration, it may not be meaningful to treat every value as a measurement of an identical species under unchanged conditions. This is a case where a number should be accompanied by a physical description of the sample. ‘A 70% yield at this concentration and geometry’ is more useful than ‘the material has 70% efficiency’ without qualification.

Distinguish absolute from relative methods

An absolute method attempts to infer emitted and absorbed photon totals through a calibrated instrument and appropriate reference configurations. A relative method compares the sample with a reference material of known quantum yield, accounting for absorbance, integrated emission, excitation and collection geometry. Relative approaches can be efficient and informative, but they inherit uncertainty from the reference and from the validity of matching optical conditions.

When comparing a relative result with a sphere-based absolute value, do not assume the word ‘quantum yield’ guarantees methodological equivalence. Identify the reference standard, solvent, refractive-index correction if applicable, optical density and emission calibration. A disagreement should prompt a comparison of assumptions before anyone accuses a laboratory of making an error. Different specimens or conditions can also genuinely produce different results.

Excitation wavelength can change the answer

Quantum yield may depend on which states are excited, how energy is transferred, whether multiple species are present, and whether the experiment enters a nonlinear regime. A single reported value must therefore be understood as a yield at stated excitation conditions. A wavelength-dependent PLQY spectrum is not the same object as an ordinary photoluminescence excitation spectrum; one asks about emission probability per absorbed photon, while the other records emission under varied illumination and a selected detection window.

A reader might be tempted to compare a blue-excited sample with a green-excited sample and call the difference intrinsic material superiority. Without controlling absorbed photon flux and sample state, that is not justified. Name the source spectrum, bandwidth, power and absorption regime. If the excitation is broadband, explain how the photon accounting was performed across its wavelengths.

Power dependence is a physical variable, not a nuisance

Some samples change efficiency as excitation power increases because nonradiative channels fill, exciton-exciton interactions become important, charge states shift, or photochemical degradation starts. A PLQY measured under weak continuous illumination need not equal a value measured under intense pulsed excitation. It is therefore unwise to take the highest available number and present it as an invariant material constant.

Power sweeps should use consistent beam geometry and report incident or absorbed photon flux, not merely a laser percentage setting. Look for reversibility after returning to the original power. If the sample’s emission does not recover, photodamage may have occurred. A published curve can still provide useful comparative information, but its conclusions should be confined to the measured intensity regime.

Temperature and environment can change radiative balance

Cooling may suppress some nonradiative pathways and improve PLQY; warming may make other channels more accessible. Oxygen, moisture, solvent polarity and solid-state packaging can alter both emission and quenching. Two values measured at different temperatures do not establish a clean comparison between two chemical formulations. They are measurements of whole sample-plus-environment systems.

Before describing a thermally activated effect, check whether the sample was stable and whether detector sensitivity or optical geometry changed with temperature. Cryostat windows can introduce reflection losses, and solvent changes can alter concentration or phase behaviour. Separate measured yield changes from proposed mechanisms. If the mechanism is tentative, explain the candidate pathways and the further evidence needed.

Quantum yield and excited-state lifetime are complementary

A time-resolved photoluminescence curve describes how an emission signal evolves following excitation. Quantum yield counts emitted photons relative to absorbed photons. In a simple single-state model, the radiative rate kr and nonradiative rate knr give ΦPL = kr/(kr + knr), while the lifetime τ = 1/(kr + knr). Taken together, yield and lifetime can estimate the rates under those model assumptions. Neither quantity alone uniquely determines both rates.

That simple model can fail for multiexponential decay, migration, delayed emission, traps, and distributions of environments. An apparently longer lifetime does not guarantee a higher quantum yield: radiative and nonradiative rates may both change. Read the companion TRPL decay-curve guide for the separate challenges of lifetime fitting, instrument response and average definitions.

Do not confuse PLQY with electronic quantum efficiency

Incident photon-to-current efficiency (IPCE) and external quantum efficiency in a photoelectric device count electrical carriers collected per incident photon under specified conventions. PLQY counts emitted light per absorbed optical photon. Both use the word ‘quantum’, but they describe different outputs and usually different denominators. A semiconductor can emit efficiently while collecting charge poorly, or collect charge effectively while emitting very little.

Ask what the measured event is. In PLQY the desired output is photons; in photoelectrochemical IPCE it is collected charge; in electroluminescence external quantum efficiency it is emitted photons per injected carrier, subject to the relevant definition. The IPCE and EQE spectrum guide illustrates why the denominator must always travel with the metric.

Use uncertainty intervals rather than unsupported precision

A PLQY of 72.341% looks authoritative, but five printed significant figures are not evidence of accuracy. Uncertainty can arise from detector calibration, lamp stability, blank subtraction, sample positioning, integration bounds, repeatability, low absorption, scattering and correction models. Some components are statistical and shrink with repeated measurement; systematic biases do not disappear simply because more spectra are recorded.

A strong report states repeat count, measured variation and the most important calibration or model uncertainties. For a set of repeated measurements, one may report a mean and an appropriate spread while explaining how instrument-scale errors were considered. If uncertainty is poorly established, rounded values and cautious comparisons are preferable. The purpose of the interval is to communicate what the data can support, not to make a claim look less impressive.

How to interpret numbers that exceed one

For an ordinary single-photon-emission process, a quantum-yield estimate above unity often signals calibration or accounting trouble, especially if a material is not known to multiply emitted photons. However, some physical processes such as quantum cutting can produce more than one emitted photon per absorbed high-energy photon, so an absolute claim that no photon yield can ever exceed one would also be wrong. The correct response is to ask which mechanism and definition the experiment supports.

Check the blank, spectral correction, detector saturation, photon-energy weighting and sample stability. If the material is genuinely proposed to show multiple-photon emission, seek independent spectral and mechanistic evidence. A surprising number is a reason to investigate, not an invitation to silently clip the graph at 100% or promote an unverified breakthrough.

A worked photon-accounting example

Suppose corrected measurements estimate 40 million excitation photons entering the relevant reference configuration and 25 million remaining unabsorbed with the sample. The estimated absorbed count is 15 million. If the sample contributes an estimated 9 million emitted photons after all appropriate background and response corrections, the illustrated PLQY is 9/15 = 0.60, or 60%. These are invented teaching numbers, not a laboratory result.

Now imagine uncertainty of roughly one million photons in the absorbed difference, combined with uncertainty in the emitted count. The nominal 60% should not be quoted as if its last decimal were known. Recompute under plausible low and high bounds and compare with actual repeat data. If two samples differ by only one percentage point, their ordering may be unresolved. The arithmetic is easy; demonstrating that both input counts are trustworthy is the harder part.

Why normalisation can destroy absolute information

Two emission curves are often scaled so that their highest peaks equal one. This helps readers compare widths, wavelengths and relative spectral features. But it erases the absolute size of the emission signal. The same normalised shape could come from a bright high-yield material or a faint low-yield one. If the excitation traces have also been normalised independently, the necessary photon balance has disappeared from the figure.

Some figures retain raw counts in supplementary data and show only normalised plots in the main text. That is acceptable if the method clearly explains how the yield was calculated from calibrated originals. It is not acceptable to infer an absolute ratio from two independently normalised graphics. Use the plot for shape and the corrected integration records for quantitative efficiency.

A caption that makes the result auditable

A careful caption might say that absolute PLQY was measured with a calibrated integrating sphere for a specified specimen, excitation wavelength and power, under stated temperature and atmosphere, with matched blanks and corrections. It should describe the specimen form, excitation geometry, spectral integration range, number of repeats and uncertainty representation. Those details tell readers what can be repeated. They also reveal whether a comparison between publications is genuinely like for like.

Avoid claiming that the spectrum alone ‘proves high stability’, ‘proves no quenching’ or ‘shows zero loss.’ Stability requires an appropriate time or stress test; the lack of a visible feature is bounded by the detection limit. PLQY is a valuable efficiency measurement but it does not replace photobleaching tests, electrical characterisation, chemical identification or structural analysis.

Interpreting the measurement in another language

Terms such as photon, radiance, irradiance, flux, quantum yield, internal efficiency and external efficiency often acquire deceptively similar equivalents across languages. Translate the complete operational definition before settling on a term. ‘Emitted’ and ‘absorbed’ cannot be interchanged; ‘incident’ is not the same as ‘absorbed’; and ‘corrected intensity’ cannot be reduced to ‘brightness’ without a caveat. The numbers and denominator should survive every language change.

When revising a scientific translation, reproduce one sample calculation and confirm that the target-language sentence states the same conditions and limits. The translation-verification guide offers a wider approach to meaning preservation, while cross-linguistic transfer explains why familiar-looking vocabulary sometimes misleads.

Worked reading: seven situations that change the conclusion

A bright powder and a dim clear film

A powder looks much brighter in a photograph than a clear polymer film. The tempting conclusion is that its PLQY is higher. First check absorption fractions, excitation spectrum, scattering and image exposure. The powder can intercept far more photons and redirect emitted light into the camera while suffering more nonradiative loss per absorbed photon. The film may absorb little and convert a larger share efficiently. A sphere experiment with geometry-matched blanks is needed to rank absolute yield. The photograph is evidence of appearance under one optical setup, not a universal photon-conversion comparison. When translating the result, say ‘appears brighter under these conditions’ until the yield measurement is available.

A denominator close to the baseline

One sample has a very small measured excitation deficit, so the instrument prints an unusually large PLQY. Before celebrating, inspect whether lamp drift during the blank measurement is similar in size to that deficit. Reposition the sample and blank, repeat at another reasonable concentration if stable, and show raw excitation and emission records. If the denominator remains uncertain, report that the yield estimate is weakly constrained. A printed value is not self-validating. The physical issue here is subtraction of large, nearly equal optical signals, not merely an algebraic mistake in the last division.

A red emission tail outside the scan

An emission spectrum is collected only to 700 nm, but other measurements indicate a broad tail beyond that wavelength. The reported yield may be underestimated because some photons were omitted. Expanding the integration window is appropriate only if the spectrometer has calibrated sensitivity there and the added counts exceed background. Otherwise the uncertainty must explicitly include the missing spectral fraction. A reader should resist simply drawing a smooth tail and counting it as measured emission. The description of the output needs to name the actual detected band and any extrapolation or exclusion.

The spectrum shifts after dilution

A dye is diluted tenfold. Its emission maximum moves, its integrated area changes and the calculated yield rises. This could reflect reduced self-absorption, aggregation changes, solvent effects or a combination. It is not enough to attribute the change to a single quenching mechanism on the basis of one spectrum. Compare absorption band shapes, time-resolved data and repeatability. Check whether the blank and cuvette remained comparable. A careful summary says that the measured yield increased with dilution while identifying proposed explanations and their remaining tests.

Two instruments disagree

A laboratory obtains 55% in one integrating sphere and 68% in another. Rather than averaging them as if both are interchangeable replicates, compare source wavelength, spectral calibration, sample positioning, blank material, integration bounds and detector response. Repeat a stable reference specimen on both instruments. If one instrument systematically differs, the discrepancy may be a traceable calibration issue. If the sample itself changed between runs, even the reference cannot resolve the entire difference. The final language should distinguish reproducibility within a system from agreement across systems.

A longer lifetime with lower PLQY

After a processing change, a film’s fitted emission lifetime grows while its quantum yield drops. This is not logically impossible. In a simple kinetic model, multiple rates determine both numbers; in a heterogeneous film, measured decay amplitudes and populations complicate the relation further. Check whether the emitting species, detection band and fit method remained comparable. An explanation should not equate ‘longer-lived’ with ‘more efficient’. Instead report each measured outcome and use an appropriate model to evaluate whether radiative rates, traps or energy transfer may have changed.

A result greater than 100 percent

An instrument outputs 106% for a routine fluorescent dye. First investigate correction settings, reference subtraction, stray excitation and denominator instability. It would be premature to market photon multiplication. If the sample is a deliberately engineered quantum-cutting material, greater-than-unity photon output might be a physically meaningful hypothesis, but it needs suitable evidence of multiexciton or multiple-photon pathways. The same numeral calls for different interpretation depending on the physics. A result should never be converted into a claim without checking how the photons were counted.

A result quoted with three decimals

A report lists 81.237%, 80.904% and 81.111% for three scans while giving no calibration uncertainty. Those differences mainly describe the repeatability of that data processing sequence. They do not show that the absolute yield is known to a thousandth of a percentage point. Ask whether spectral correction and systematic errors have been evaluated. A rounded mean with a candid uncertainty statement can convey more genuine scientific information than the unqualified display of all digits.

A changed substrate and unchanged emitter

The same luminescent coating is placed on a dark substrate and a reflective substrate. PLQY calculations differ. One possibility is changed optical escape and reabsorption; another is genuine interfacial quenching or a change in the film’s microstructure. Compare optical backgrounds, excitation geometry and time-resolved responses. A difference in measured external emission is not automatically a difference in the intrinsic radiative probability of an isolated molecule. Language should preserve which physical system was tested: film alone, film-on-substrate, or a device stack.

A broad excitation lamp

A white-light source is used, and the methods vaguely state that the emission area was divided by the absorbed power. Because photon energy varies with wavelength, that division may not be a valid quantum-yield calculation. Ask how the incident spectral photon flux was characterised, how the absorbed fraction was determined and whether the emission count was wavelength corrected. A broadband measurement can be valid with appropriate bookkeeping, but a one-number optical-power ratio is insufficient evidence by itself. The crucial issue is defining the denominator over the source spectrum.

A short decision route for readers and reviewers

First identify whether the plotted quantity is a corrected photon ratio, raw emission intensity or some other optical measurement. Then check excitation conditions, the blank and sample geometry, spectral calibration and the absorbed-photon calculation. Inspect the range used to integrate emission, and ask whether self-absorption and scattering require explicit treatment. Review replicate measurements and the uncertainty budget. Only after those checks should you compare materials or connect the measured yield to a mechanism.

The most useful outcome is a sentence that another scientist could try to falsify: ‘Under the specified excitation, sample geometry and calibration procedure, this specimen emitted an estimated fraction of the photons it absorbed.’ That statement is modest by design. It is still powerful because it names what was measured and leaves a clear path for checking it.

How to Translate Easily | How to Read Photoluminescence Quantum Yield (PLQY) Measurements, Integrating-Sphere Data and Corrections