Until recently, many physicists thought it would be impossible to image objects with unrestricted resolution because of the so-called diffraction limit, which prevents objects that are much smaller than the probing light's wavelength from being seen. However, techniques such as near-field imaging have been able to beat this limit by using a tiny light source that can get to within a wavelength of the surface of a sample. In this way, sub-wavelength surface features reveal themselves by reducing the transmitted intensity of the oncoming light. But so far there have been no light sources that can acquire sub-wavelength images from biological samples, which often need to be supported in warm, wet conditions.

Now, however, a group led by Peidong Yang at the Lawrence Berkeley National Laboratory in the US has discovered that nanowires made from potassium niobate can be used as optical light sources for sub-wavelength imaging. They began by dispersing many nanowires into water, and then used an infrared optical tweezer – a laser that can apply a force to nano-sized objects – to grab just one.

A unique feature of the potassium-niobate nanowires is that they can absorb two photons at a time from the optical tweezers and then reemit one at twice the frequency of the originals. This "frequency doubling" is useful because it produces green light that would be able to effectively highlight parts of biological samples coated with fluorescent dye – a common "tagging" method used in biological imaging.

Yang's group used the optical tweezer to scan the nanowire over the surface of a thin sheet of glass imprinted with 50-nm thick lines of gold, and recorded the intensity of the reemitted light transmitted onto a detector. As the nanowire passed over each gold line, the intensity dropped, allowing them build up an image of the surface.

Because the nanowire light source needs no wires, it can be operated in liquids safely, making it ideal for biological samples. "We get many neat ideas from the physics community that fall flat when they are applied to biology," Warren Zipfel, a biomedical engineer at Cornell University in the US, told Physics Web. "If this can really operate [in biological conditions], that's a big plus."

The US group says the next step is to refine the signal-processing techniques to make the imaging technique as practical as others such as atomic force microscopy.