Science

Ultrathin microscope gets images faster

Ultrathin microscope gets images faster
Scientists have created a thin handheld microscope that can obtain high-quality images in a fraction of the time required by traditional scanning microscopes (Photo: Fraunhofer)
Scientists have created a thin handheld microscope that can obtain high-quality images in a fraction of the time required by traditional scanning microscopes (Photo: Fraunhofer)
View 1 Image
Scientists have created a thin handheld microscope that can obtain high-quality images in a fraction of the time required by traditional scanning microscopes (Photo: Fraunhofer)
1/1
Scientists have created a thin handheld microscope that can obtain high-quality images in a fraction of the time required by traditional scanning microscopes (Photo: Fraunhofer)

With conventional microscopy, if a scientist wishes to obtain a high-resolution image of a relatively broad area, they typically have to use a microscope that scans across that area in a grid pattern, recording many images one point at a time. Those images are then joined together to form one complete picture. Such systems take a long time to perform a scan, so both the microscope and the subject must be held still while it's taking place. Researchers from Germany's Fraunhofer Institute for Applied Optics and Precision Engineering, however, have created a thin, handheld microscope that can reportedly obtain similar-quality images in less than one second.

Unlike a scanning microscope, that records many single images one after the other, the Fraunhofer microscope uses an array of tiny lenses to record a comparable number of images all at once. As with the scanning microscope, these are subsequently combined to form one complete image.

The new microscope's imaging system consists of three glass plates, stacked one on top of the the other like pancakes. Each plate is covered with a matrix of the tiny lenses, both on its top and bottom surfaces. Looking down through the plates from above, each tiny lens lines up both with its counterpart on the other side of its plate, and with the other lenses that occupy the same location on the other plates. Microscopic details are therefore imaged through a stack of six tiny lenses, along with two achromatic lenses. These stacks of lenses are called channels, and it is the images produced by the multiple channels that are digitally joined together, side-to-side and top-to-bottom, to create the complete picture.

Because it has an optical length of just 5.3 millimeters, the microscope is able to maintain a very flat profile.

To make the lenses, the scientists start by coating a glass plate with photoresistant emulsion, covering it with a mask in the pattern of the lens matrix, then exposing it to UV light. Emulsion exposed to the light hardens, while the emulsion protected by the mask washes away when exposed to a special solution. This leaves a matrix of tiny cylinders, which are then heated. This causes them to partially melt, and form into spherical lenses. The lens-covered plate is then used to create a die, which in turn can be used for mass production - glass substrates are coated with a clear liquid polymer, the lens die is pressed down into that, the polymer takes on the shape of the lens array, and is then hardened using UV light.

The microscope is currently still in the prototype stage, and probably won't go into production for at least one or two years. Once it does, it could be used to examine suspicious skin blemishes, check documents for authenticity, or various other applications. It is currently capable of imaging of objects the size of a matchbox, in one pass.

1 comment
1 comment
attoman
The use of microlenses to make large area images by reduced range scanning (not stated but implied by the extremely long image time which should be less then 1/30th of a second for non-scanners) is decades old and was made by a local Bay Area man Marc Davidson.
Since no N.A. was stated it is difficult to know if the present limits of about .18 N.A. for such systems have been improved. The goal in such a device is an N.A. of at least .7 to permit useful medical and industrial uses and the outstanding real problem of such an implementation is not mentioned in this article at all.
Perhaps Fraunhofer hasn't figured it out yet.