Handheld NIR, Peer-Reviewed

A 29-Page Answer to One Question

Portable near-infrared spectroscopy is a category that has grown faster than it has been examined. Marketing claims come in a predictable shape: the laboratory in the palm of your hand, instant insight, nothing lost in miniaturisation.

A useful counterweight is a peer-reviewed review, and one appeared in 2023 — Handheld Near-Infrared Spectroscopy: State-of-the-Art Instrumentation and Applications in Material Identification, Food Authentication, and Environmental Investigations, published in Chemosensors (11, 272), co-authored by Prof. Dr. Heinz Siesler and Frank Pfeifer of the University of Duisburg-Essen. It is worth reading not for what it says about any particular instrument, but for the shape of the evidence it assembles. Three points stand out.

Six Ways to Build a Portable

The review opens by naming the pattern it does not want to repeat: “the exaggerated and sometimes unrealistic narrative of some direct-to-consumer companies,” which it says has “raised unrealistic expectations” for a valuable technology. A fair warning, and one our copy should never need to answer for.

The review then maps the field into six instrument families, distinguished by how they split the light: linear variable filter (LVF) instruments, digital micromirror (DMD) devices, Fabry–Pérot tunable filters, micro-Fourier-transform (MEMS FT) spectrometers, grating microspectrometers, and one bank-of-organic-solar-cells scanner design.

The map matters because it explains the catalogue of portable instruments on the market: each family inherits its own strengths, its own limits, and its own failure modes. “Portable spectrometer” is not one technology, and comparing instruments across families using a single number (sensitivity, speed, size) usually misleads.

The Trade-Off Nobody Admits

The review’s most honest finding is structural: signal-to-noise ratio, scan speed, and instrument stability sit on one axis. Integrating longer raises the signal-to-noise ratio but slows the measurement; pushing scan speed has the mirror-image cost; and thermal and mechanical stability is what quietly turns a good reading into a repeatable one.

There is no free lunch in the physics. An instrument can be small, fast, and quiet all at once only if one of those claims is softer than it sounds. The review’s benchmark studies are the evidence base for this — and the benchmark numbers are strong where the science is well established.

Where Portable Has Earned Its Keep

The evidence is strongest in food and agriculture, where the chemometrics problems are mature:

• Almond flour authenticity. Three different portable instruments identified adulteration with 100% sensitivity and greater than 95% specificity at adulterant levels of 5% (w/w) or above, with PLS models above R² = 0.90.
• Quinoa flour. A portable spectrometer detected adulteration across 0–51% (w/w) with a predictive R² of 0.98 — a working purity model, not a demo.
• Soil oil contamination. A portable engine quantified hydrocarbon contamination in soil across a 0–12 % (w/w) range against laboratory references — an in-field result, not a demo.

Two caveats from the same review deserve the same prominence. First, heterogeneous bulk solid is the hard case — which is why good portable practice for powder and bulk includes a rotating sample turntable (the review documents designs measuring up to 120 points per sample) rather than a single-spot reading. Second, the review flags a specific hype pattern: the push to move evaluation into cloud-based big-data pipelines before the underlying spectroscopic science has been validated.

Our Position on the Cloud

We find that last point well-taken, and it sits at the heart of how we build. The cloud in a portable spectroscopy workflow should do what the laboratory already uses it for — validation, model comparison, and reporting at scale — not what no cloud does by itself: take a better measurement. The measurement is made by the spectrometer, at the point of need, against the physics. Everything upstream of that reading is engineering; everything downstream is interpretation.

A Note on This Publication

Altum Solutions is named in the acknowledgement of the Chemosensors review as a supplier of sample sets and reference values to the environmental section. We mention it plainly: it is a sample-supply credit, not an authorship claim, and the review’s findings stand on its own references.

Full reference: Yan, H.; De Gea Neves, M.; Noda, I.; Guedes, G.M.; Silva Ferreira, A.C.; Pfeifer, F.; Chen, X.; Siesler, H.W. “Handheld Near-Infrared Spectroscopy: State-of-the-Art Instrumentation and Applications in Material Identification, Food Authentication, and Environmental Investigations.” Chemosensors 2023, 11 (5), 272. https://doi.org/10.3390/chemosensors11050272 (open access, CC BY).

Next in the NIR Academy: the instrument-side paper — why the spectrometer engine is the part of the product that actually decides performance. See also Wetware-in-the-Loop Testing.