Three dimensional touch and vision for the micro-world

The ability to observe at tiny length scales has enabled key advances across the physical and life sciences. Much of what we know about the structure of cells and tissues comes from experiments on the micron length scale, enabled by new microscopy techniques. Modern manufacturing is increasingly concerned with materials that are structured on the nanometre scale, and devices which have ever-smaller features. Manipulating and measuring microscopic objects is a problem common to fields as diverse as microfabrication and cell biology, and it is these challenges that my doctoral studies have addressed. Tiny sizes mean tiny forces; so small that the light from a laser can be used to propel objects. Optical tweezers, a technique pioneered some two and a half decades ago, exploit light’s momentum to trap and manipulate objects. Now an established tool, single particles can be trapped and tracked to measure forces on a molecular scale, and this work is responsible for much of our current knowledge of motor proteins. This thesis describes advances in the holographic technology used to control multiple optical traps (and hence many trapped particles), and improved methods for monitoring the positions and forces involved. The speed with which multiple holographic optical traps can be moved has traditionally been limited by the time taken to calculate holograms, but by using consumer graphics cards and high speed Spatial Light Modulators (SLMs) I have implemented holographic systems fast enough to react to the Brownian motion of trapped particles. Brownian motion can, to some extent, be suppressed by this approach, and it also allows the trap's stiffness to be engineered to balance sensitivity against tight constraint of position. Feedback control using an SLM, rather than the other beam steering technologies that have been employed, is able to react to motion in three dimensions. This requires 3D position measurement, which is provided by the stereo microscopy technique described in Chapter 2. By illuminating and viewing the sample from two different angles it is possible to reconstruct the depth of objects. This is accomplished through a single high numerical aperture microscope objective, the same lens used to focus the trapping laser. In conjunction with a fast CMOS camera, it is possible to track particles with an accuracy of 2-3nm at several thousand frames per second. This allows measurement of forces and displacements within the control loop, that can be fed back to influence the position of the optical traps. This force information can also be relayed to the operator using a force-feedback joystick as detailed in Chapter 7. Interface design is an important part of making technology accessible to scientists from other disciplines; to this end I have also developed a multi-touch tablet application to control optical tweezers. By creating simple, reliable systems and coupling them to an intuitive interface, I have endeavoured to produce developments which are of use to the non specialist as well as to experts in optical tweezers-a number of which are now available commercially (Section 8.7). These technologies form the basis of a toolkit for working with multi-part probes in optical tweezers, and they should bear fruit in the coming years as a new form of scanning-probe microscopy emerges