High-speed feedback trap based on optical tweezers

我和我的团队提议建立一个基于光镊的高速三维反馈陷阱，以研究统计物理学和生物物理学的基本问题。在过去的十年里，我们开发了一种使用反馈陷阱来定量研究被困在二维“虚势”中的粒子运动的方法，这些“虚势”可以被任意操纵。我们使用反馈陷阱来进行非平衡统计物理的基本测试，重点关注将热力学与信息论联系起来的问题。两个亮点包括兰道尔原理的测试，该原理指出，删除一个比特的信息不可避免地需要至少kT ln 2的功，以及非平衡系统熵的函数形式的第一次测量。******使用2014年RTI奖资助开发的仪器，我们展示了光学镊子可以提供制造反馈陷阱所需的力。最新版本的陷阱速度比最初的设计提高了1000倍，空间规模减少了100倍。我们还展示了一种基于虚拟双阱电位的双态胶体系统，该系统在10nm的空间尺度上具有近1khz的跃迁速率。最后，对于简单的捕获，反馈可以使捕获的有效刚度增加40倍，而光束强度没有增加。这项仪器工作刚刚发表在《应用物理快报》上，并在美国物理研究所的Scilights网站上进行了重点介绍。******到目前为止，主动反馈操作仅限于阱中粒子的横向（xy）运动。轴向(z)方向的捕获仅限于反馈镊子提供的被动捕获。不幸的是，梁的几何形状（和辐射压力）导致轴向刚度明显低于横向刚度。在提出的设置中，我们将改进传感器和执行器的设计，以增加垂直运动的主动控制。进一步的改进将使整体反馈带宽增加10到1 MHz。******增加三维控制和进一步加速陷阱带宽将导致定性的新类型的应用。第一个目标是将随机热力学扩展到基于轨迹的量，这是由于缺乏统计数据而在其他实验中未探索的细节水平。第二个目标是操纵不稳定的物体，如反馈陷阱中的气泡。第三个目标是重新优化光镊的参数，以充分利用反馈控制提供的可能性，这是迄今为止不可能实现的壮举，因为反馈控制需要对所有三个轴进行控制。因此，我们提议建造的仪器将使统计物理学的新基本测试和生物物理学的新测量技术成为可能。*****

My group and I propose to build a high-speed, three-dimensional feedback trap based on optical tweezers, to study fundamental problems in statistical physics and biophysics. Over the past ten years, we have developed a way to use feedback traps to quantitatively study motion of particles trapped in two-dimensional "virtual potentials" that may be arbitrarily manipulated. We have used feedback traps to carry out fundamental tests of nonequilibrium statistical physics, focusing on questions connecting thermodynamics with information theory. Two highlights include a test of Landauer's principle, which states that the erasure of one bit of information inevitably requires work of at least kT ln 2 per bit and the first measurement of the functional form of nonequilibrium system entropy.******Using an instrument developed with funding from a 2014 RTI award, we showed that optical tweezers can provide the force needed to make a feedback trap. The latest version of our trap has demonstrated a speed-up of 1000 times relative to our original design and a decrease in spatial scale by a factor of 100. We have also demonstrated a two-state colloidal system based on a virtual double-well potential that had nearly 1 kHz transition rates at a spatial scale of 10 nm. Finally, for simple trapping, feedback can increase the effective stiffness of the trap by a factor of 40, with no increase in beam intensity. This instrumentation work has just appeared in Applied Physics Letters, where it was highlighted on the Scilights website of the American Institute of Physics.******To date, active feedback manipulation has been confined to lateral (xy) motions of particles in the trap. Trapping in the axial (z) direction was limited to the passive trap provided by feedback tweezers. Unfortunately, the beam geometry (and radiation pressure) leads to axial stiffnesses that are significantly lower than the lateral ones. In the proposed setup, we will improve sensor and actuator design to add active control of vertical motions, as well. Further improvements will increase the overall feedback bandwidth by 10 to 1 MHz.******Adding three-dimensional control and further speeding up the trap bandwidth will lead to qualitatively new types of applications. A first goal is to extend stochastic thermodynamics to trajectory-based quantities, a level of detail that is unexplored in other experiments for lack of statistics. A second goal is the manipulation of unstable objects such as air bubbles in a feedback trap. A third goal is to re-optimize the parameters of optical tweezers to take full advantage of the possibilities offered by feedback control, a feat that has not been possible so far, as feedback control needs to be for all three axes. The instrument that we propose to build will thus make possible new fundamental tests of statistical physics and also new measurement techniques in biophysics.*****