If laser flash is the aristocrat of bulk thermal measurement — demanding machined discs and coatings — the transient plane source — often implemented in commercial Hot Disk® systems — is the pragmatist. Press a thin double-spiral sensor between two halves of almost anything solid, pass a current step, and record how the sensor’s own temperature climbs: within seconds the analysis returns both thermal conductivity and thermal diffusivity, with no disc machining, no optical coating, and validated accuracy across four orders of magnitude in conductivity12. That reach made the Hot Disk configuration a laboratory staple. This article explains how one spiral does two jobs, why a time window — not the instrument — decides whether the number is valid, and where the clamped-halves architecture ends.
- 1How one spiral does two jobs
- 2The probing depth: a clock that must stop twice
- 3Interactive: watch the legal window open and vanish
- 4Where TPS earns its ubiquity
- 5Where the clamped-halves architecture ends
- 6TPS in the method landscape
- 7Frequently asked questions
- 8Keep exploring the knowledge hub
- 9References
How one spiral does two jobs
Gustafsson’s 1991 formulation crystallized the design: a nickel double spiral, photoetched and clad in thin insulation, acting simultaneously as a plane heat source and a resistance thermometer1. Clamp it between two halves of the sample, apply a constant power step, and the spiral’s resistance — hence its mean temperature — is sampled continuously. The mathematics of a disc-shaped source in an infinite medium then does something a single-property method cannot: because the shape of the temperature rise depends on diffusivity while its scale depends on conductivity, one transient constrains both, with volumetric heat capacity following as the quotient13.
Validation work in the following years established the credentials that matter to a working laboratory: agreement with the specific reference materials of the cited validation studies within about five percent, across a span from insulating polymers to metals — a validated-condition result, not a blanket platform accuracy2, and extensions to thin samples using constrained-geometry modules3. The preparation economics explain the rest of the adoption story: two reasonably flat faces are the headline sample requirement — with thickness and lateral size relative to the sensor radius, homogeneity, contact pressure and the power/time window all standing behind it in the fine print.
The probing depth: a clock that must stop twice
Everything rigorous about TPS lives in one quantity: the probing depth, d = 2√(αt), the radius to which the thermal disturbance has meaningfully spread after time t1. The infinite-medium solution is honest only while that radius stays inside the actual sample. So the analysis clock must stop twice. It must start after the earliest data — where the sensor’s own heat capacity and the sensor–sample contact dominate the record — and it must stop before the probing depth reaches the sample boundary, after which the fitted conductivity silently absorbs the error of a violated model12.
The window is therefore a property of the specimen, not the instrument. A thick insulator opens a generous window; a thin, fast-conducting plate can close it entirely — boundary reached before the contact transient clears — which is precisely the regime where the thin-sample modules, and beyond them the free-standing-sample methods, take over34. A TPS number reported without its analysis window is as incomplete as a flash number without its pulse correction5, and window-selection sensitivity deserves the same regression scrutiny as any fit range choice67 — the same finite-sample physics wearing a different clock.
Interactive: watch the legal window open and vanish
The simulator draws the probing depth climbing as √t against your sample’s boundary. Dial diffusivity up and thickness down and watch the legal window — after the contact transient, before boundary arrival — shrink to nothing: the visual definition of a sample that has exited the standard TPS architecture.
Where TPS earns its ubiquity
Three structural wins. Preparation economics: no machining to a disc, no graphite coating, no lithography — two flat faces suffice, which is why TPS often produces the first number on a new material while other methods are still preparing samples1. Two properties per transient: conductivity and diffusivity from one record, hence volumetric heat capacity as well — closing the k = αρcp triangle in a single clamping — with the caveat that the quotient inherits both parents’ uncertainties, bookkeeping the standard frameworks formalize389. Dynamic range: the validated span from insulating foams to metals covers most engineering solids in one architecture2. For anisotropic materials, dedicated analysis modes separate radial and axial transport under stated assumptions — a capability, with the assumptions doing real work that the report must disclose10.
Where the clamped-halves architecture ends
Contact is the price of convenience. The sensor touches the sample through its insulation layer, and imperfect contact appears as an added interface resistance concentrated in the early transient. The standard defense — discard early data — costs window length, and on rough or rigid samples the residual contact term remains the leading systematic; contact physics is a first-class subject for good reason11.
Sample classes that never clamp. A single fiber, a wire, a free-standing film strip, a micro-scale specimen: nothing about two clamped halves applies. These are the suspended electrothermal family’s native geometry — micro/nanoscale conductors measured as their own sensing elements12 — the sample itself carrying the heating and sensing that TPS delegates to its spiral413 — with nano-objects belonging further down to microfabricated bridge devices1415.
Sub-micron films and interfaces. The spiral’s millimeter scale cannot resolve nanometer layers or single interfaces; those jurisdictions belong to 3ω and thermoreflectance1617, with interfacial conductance a measured quantity only in the latter family18.
TPS in the method landscape
| Question | TPS1 | LFA19 | 3ω16 | TET family4 |
|---|---|---|---|---|
| Preparation | Two flat faces | Machined disc + coating | Lithography | Mount + silver paste20 |
| Properties per run | k and α together3 | α (k via ρcp) | k | α (k, ρcp via companions21) |
| Validated span | ~4 decades in k2 | Bulk solids, extreme T5 | Dielectrics, films22 | Fibers, films, micro-samples13 |
| Leading systematic | Sensor contact + window discipline | Pulse + heat-loss corrections23 | Geometry idealization | Radiation/coating, audited by differentials24 |
| Fibers / wires | No | No | No | Native25 |
Frequently asked questions
How flat do my sample faces really need to be?
Flat enough that the sensor’s insulation touches across its spiral area without air gaps a thermal wave would notice. Machining-grade flatness is unnecessary; visible waviness or grit is disqualifying. Early-time residuals catch many contact problems — but not all: a contact resistance can also correlate with the fitted parameters and produce a smooth, plausible, biased result, which is why reference-material checks remain part of honest practice111.
Can TPS measure liquids and pastes?
Configurations exist for low-viscosity media — with an appropriate cell, and a measurement window kept short enough to suppress convection and sedimentation — the probing-depth logic carries over, with convection replacing the boundary as the window-closing enemy. Free convection onset, not the sensor, sets the practical limit1.
My plate is 0.4 mm thick and conducts well — TPS or something else?
Run the window arithmetic first: if boundary arrival precedes the contact transient’s clearing, the standard architecture has nothing valid to fit. Thin-sample modules extend the reach under added assumptions3; a free-standing strip of the same material measures natively in the suspended family instead4.
Keep Exploring the ACS Thermal Metrology Knowledge Hub
This article is one chapter of the ACS thermal metrology knowledge hub. To keep going:
- Thermal conductivity & diffusivity testing: the pillar guide — the full landscape in one place.
- Which thermal measurement method should you use? — the interactive decision hub.
- Laser flash analysis explained — the disc-shaped alternative and its corrections.
- What is thermal conductivity? — the k–α–ρcp triangle TPS closes in one run.
- Thermal testing services — send the sample; get defensible numbers back.
References
This article describes the transient plane source method and its comparison with flash, 3ω and suspended electro-thermal techniques for educational purposes. The interactive model is a normalized schematic of the probing-depth window, not the instrument’s residual analysis. For sample-specific feasibility and formal quotes, contact our thermal testing team.