Development of a Nano-Illumination Microscope

Abstract

This doctoral thesis proposes and explores a new approach to lensless microscopy, focusing on making high resolution imaging ubiquitous and low cost. A short introduction to microscopy frames the state of current techniques: Abbe’s law limits the resolving power for visible light microscopes with lenses, techniques using UV, X-rays, or electrons are incompatible with live samples and all of them, including super-resolution microscopy methods, are complex devices not suitable for being used in the field as mobile devices. Some lensless microscopy methods try to solve these issues.

The microscopy method is named Nano Illumination Microscopy (NIM) because it relies on using nanometric light sources in an ordered array to illuminate a sample placed in close proximity to them, and a photodetector at the other side to measure the amount of light arriving from each LED. In a setup like this, the resolving power is provided by the nano-LEDs and their distribution instead of the sensing devices, as is the case in the other methods. Since the resolving power depends on the pitch of the LED array, this method also opens a path to obtain super-resolution images, depending only on obtaining LED arrays with pitches smaller than Abbe’s limit for the wavelength.

After the introduction to microscopy setting the context of the thesis, the thesis continues explaining the main components used to build the microscope: a SPAD camera, designed within the context of this work, and the electronics to control the nano-LED array.

The third chapter of this thesis provides an overview of the microscopy method and its fundaments, exploring the requirements and capabilities. Image formation is first introduced with simulations, and this information is then used to build the very first prototype, a microscope capable of forming 8x8 pixel images -since that is the form factor of the LED array used, with LEDs of 5 μm in size (and 10 μm in pitch). The first results from this technique are presented and compared with the simulations, showing the agreement between both, validating the method, and offering insight on building the next prototypes, which will use smaller LEDs in an attempt to study the technological limits.

The thesis continues with the work done in search of the limits of the technique, building and testing new improved versions of the microscope and confronting the limitations which arise. Some of those came from the structure of the LED arrays themselves: while nano-LEDs well below the sizes used have been reported, those have been isolated structures or non individually addressable. Selecting exactly which LED will emit is one of the main problems to solve since with increasingly large arrays, the connections required increase exponentially until routing is impossible. The thesis also studies this problem, as the LED arrays were changed in search of the proper solution. This implied moving from a direct addressing strategy, in which each LED was selected individually, towards a matrix-addressing format, in which the LEDs are selected by polarising the appropriate row and columns.

The microscopy technique is validated and the more advanced prototypes presented. Images with a maximum resolving power of 800 nm are shown, and the results discussed, since the physical limitations on fabricating the chips limit the maximum resolving power below what was theoretically expected. The thesis also offers a short overview into the future of the Nano Illumination Microscopy technique.

Aquesta tesi doctoral proposa i explora una nova aproximació a la microscopia sense lents, amb la intenció de facilitar l’obtenció d’imatges d’alta resolució amb baix cost i disponible arreu. S’ha batejat aquest mètode de microscòpia com a Microscopia de Nano-Il·luminació (MNI) perquè la imatge es construeix a partir de fonts de llum de mida nanomètrica distribuïdes en una matriu que il·luminen la mostra de forma consecutiva i ordenada. Un sensor a l’altre costat recull la intensitat de llum que arriba de cada LED, creant un mapa de l’objecte observat. Aquest mètode fa que la resolució de les imatges depengui de la mida i distribució dels LEDs, en comptes de la del sensor com és el cas convencionalment, obrint la porta a noves integracions.

En la tesi s’ofereix una introducció general a la microscòpia abans d’entrar a detallar els components del microscopi i com s’integren per muntar-lo. A continuació es presenta i s’estudia el funcionament del mètode, començant amb simulacions i seguint amb la construcció del primer prototip de microscopi amb el que s’obtenen les primeres imatges.

La tesi procedeix a continuació a investigar els límits actuals de la tècnica de microscòpia, utilitzant noves versions de la matriu de LEDs i estratègies alternatives per intentar superar-ne les complicacions tècniques. Així, s’obtenen imatges amb una resolució de 800 nm i es discuteix la problemàtica d’implementar dispositius que s’aproximin a les expectatives teòriques per la tècnica.