A microscope that catches the slightest hints of heat

A superconducting transition-edge sensor (TES) is a device well-known for its extreme sensitivity to photons, the particles of light — so much so that they can count photons one by one. They also have very little noise, which makes their readings quite reliable. TESs are often used in single-photon detectors in quantum communications systems and in cryogenic bolometers (devices that measure infrared radiation) in astronomy. But for these virtues, however, engineers haven’t been able to use TES technology together with scanning-probe optics, where scientists use a physical probe to image surfaces at extremely high resolution. In atomic force microscopy, for example, a very sharp tip is mounted on a flexible cantilever over a surface to measure forces between the tip and the sample at the nanoscale. This technology gap has been important to fill because scanning-probe optics are currently limited by how sensitive detectors are to light fields just a few nanometres big. In other words, the missing piece was a device that married the sensitivity of a TES device with the ability of a scanning probe to access spatial scales of nanometres. A new effort by researchers from Singapore, Switzerland, and the US has offered to fill this gap using a bespoke new technique called bolometric superconducting optical nanoscopy (BOSON). According to the researchers, BOSON integrates a superconducting TES directly into a scanning near-field optical microscope. The findings were published in Physical Review X on July 25.

‘Near field’ has a simple meaning. In conventional microscopy, like the simple light microscope in a high-school biology lab, light from a sample is captured through lenses and eventually sent to the eyes of the observer. This is called far-field microscopy because the light that contains information about the sample under study travels several multiples of its wavelength before interacting with the detecting elements. In near-field microscopy, light travels much less than a single multiple of its wavelength before reaching these elements. For example, if the wavelength of the light is 500 nm, it may travel 5 cm — or 100,000-times its wavelength — before striking the lens. On the other hand, near-field microscopy, also called near-field nanoscopy, captures and analyses light that has travelled much less than 500 nm from the sample. Devices of this kind routinely use junctions made of graphene, semiconductors or metals to translate the properties of the light energy into a measurable electrical current. These technologies demand high optical power, in the milliwatt to sub-milliwatt range, as well as elaborate engineering. They also struggle to detect changes in a sample that produce weak electromagnetic fields, like vibrating atoms in some crystals. Graphene-based devices that reveal temperature changes in a sample by shifting their resistivity are also limited by the fact that graphene’s resistivity changes very weakly with temparature, limiting the devices' usefulness in bolometry. The team behind the new study thus set about looking for a detector whose resistance would change abruptly with even a small thermal load. This was BOSON.

At the heart of BOSON is a bridge. It’s made of niobium, a metal that becomes a superconductor at very low temperature. It’s also only 200-250 nm wide, a really small size that makes it extremely sensitive to heat. Imagine a single snowflake landing on your finger: even the gentle heat from your body suffices to melt it quickly. Similarly, even a small amount of heat will cause the niobium bridge’s temperature to rise enough to jerk it out of its superconducting state. The bridge sits between wider niobium leads. At the start of the researchers’ experiment, the team passed a constant current through the bridge. Hovering just above the bridge was the small, sharp tip of an atomic force microscope. When an infrared laser struck the probe tip, it concentrated the electromagnetic field onto the bridge. When the tip-induced field raised the electrons' temperature by only a few millikelvin, a "hot spot” formed on the niobium bridge. In this region, the bridge resisted the flow of current enough for a voltage to register between the leads at the ends of the bridge. This voltage was the ultimate signal of interest, demonstrating that BOSON could reliably detect extremely small changes in temperature.

The researchers also found that BOSON’s resolution is limited not by the size of the atomic force microscope’s tip (around 20 nm tip) but by the lengths across which the energy diffuses into the bridge — under 1 micrometre in the niobium bridge — and the size of the bridge itself. The researchers have written that further narrowing the bridge could further improve its spatial resolution.

Still, to highlight BOSON’s optical reach in their study, they overlaid the bridge with a 50-nm thick flake of hexagonal boron nitride (hBN), a material known to contain an unusual kind of wave called hyperbolic phonon-polaritons when illuminated with mid-infrared light. Hyperbolic phonon-polaritons are formed from when photons interact strongly with vibrations in the grid of atoms in a crystal, especially when the vibrations are within a particular frequency range. This interaction allows light to be guided into tracks that are narrower than the diffraction limit — a very desirable ability in microscopes trying to achieve a high resolution. The team shone an infrared laser at the hBN crystal to produce hyperbolic phonon-polaritons, then monitored the niobium bridge. They found that the phonon-polaritons produced an electromagnetic field in the crystal and the bridge was sensitive to changes in this field even when the latter’s power was as feeble as 50 nanowatt — fully four orders of magnitude below the power required to draw the attention of existing near-field microscopes. According to the researchers, this dramatic advance stemmed from operating the detector exactly at its superconducting transition temperature, where the bridge’s sensitivity to temperature changes is highest. BOSON also revealed how the phonon-polaritons dispersed within the hBN crystal, found to be consistent with theoretical predictions. The team said that since the bridge width is the effective detector size, future bridges that are only tens of nanometres wide should be able to study materials like hBN with even more sensitivity.

By combining a superconducting bolometer with a scanning probe, the team has shown that BOSON is a universal, cryogenic nano-optical detector whose sensitivity rivals the best available TES devices. The platform can reportedly detect weak shifts in the energy of a material with nanometre precision while depositing a negligible amount of energy into the sample, a feature that could prove useful in the study of quantum materials, which are typically very fragile. According to the team’s paper, an improved BOSON may in future may be able to detect single polaritons (quasiparticles each made of a photon coupled to an electric dipole) and be sensitive to electromagnetic fields with ultra-high frequencies (in the terahertz range). They’ve also speculated that thinner superconducting bridges and the use of improved techniques to sense voltage across them could make BOSON sensitive to power changes even slighter than nanowatts.

Featured image: A schematic diagram of the experimental setup of BOSON. CP refers to ‘Cooper pairs’, which are the charge carriers in a superconductor. I_bias is the biasing current applied to the niobium bridge. Credit: Phys. Rev. X 15, 031027.