MICROMANIPULATION

Micromanipulation is important to several DOE missions. The RIM Roadmap describes the importance of developing micro-electromechanical systems (MEMS) for the next generation of weapons systems. Many of the MEMS require automated systems for microassembly. The DOE Office of Biological and Environmental Research develops new technologies for the medical sciences. In the future, micromanipulation systems will be developed to perform surgery on the components of a living cell. Miniaturization of microchemical systems is driven by the opportunity to revolutionize heat exchangers, heat pumps, combustors, gas absorbers, solvent extractors, fuel processors, and other similar systems [1]. For energy storage and chemical processing devices there are three fundamental advantages for miniaturization. First, micro chemical processing systems have the potential for distributed and mobile applications. Second, microscale systems for heat and/or mass transport are very efficient. Third, mass production may lead to reductions in production costs similar to the reductions in the electronics industry. In addition to DOE, many other organizations will benefit from miniaturization. Jerman [2] described the medical industry's interest in miniaturization of medical dispensing technologies ranging from medical valves to advanced PCR devices. Rudolph [3] described the defense industry's interest in miniaturization of information gathering devices.

Two of the tasks in the Computational Nanoscience section of this proposal require nanomanipulation. Manipulation could be used to place quantum dots with nanometer resolution in an ordered pattern. If the quantum dots are arranged by either using DNA for directed self assembly or nanomanipulation, a scanning microscope with nanometer resolution could be used to measure the positions of the quantum dots and the electrodes or to measure the charge on each quantum dot. A scanning microscope with nanometer resolution could be used to measure the forces that are required to slide a nano object on a fixed surface that is dry or coated with a fluid film and is being vibrated at different frequencies and amplitudes.

Objective

We propose to develop intelligent control systems for two paradigm problems, micro and nano manipulation and automated assembly of micro machines. Our three year goals are to: (1) insert a submicron bead into a cell and map the inside of the cell, and (2) develop a parallel assembly system for parts that range in size from 50 to 1000 microns with sub-micron details and clearances.

Background

Manipulation systems that work in the micro world (with components and tolerances that range from nanometers to millimeters) must be based on new design principles. In the macro world we live in, gravity is a dominant force. In the micro world, surface forces (including electrostatic, van der Waals, and surface tension) are larger than gravity. In the macro world, manipulation requires a gripper and a grasping strategy. In the micro world, the part may jump toward the manipulator, no gripper may be needed, and the challenge may be to release the part. Israelachvili [4] provides a comprehensive introduction to surface forces.

Although space limitations preclude a comprehensive literature review, we will discuss the results from a few key papers. The Scanning Electron Microscope (SEM) can image micron-sized objects but creates charges on the objects that cause large electrostatic forces. Saito [5] performed pick and place operations with 2 mm polystyrene spheres in a SEM environment with high precision (errors less than 140nm).). At higher magnification, greater exposure to the electron beam (EB) results in more charge and more force. His basic strategy for pick and place operations is to avoid long exposure at high magnification to the EB. His strategy is to perform pick and place by kinematic motions without visual feedback from the SEM. For a pick operation, the adhesion between the sphere and the surface is higher than the adhesion between the sphere and the tool because the sphere has been observed and exposed to the EB. The pick strategy is to push on the spherical object until it begins to roll and then lift the tool. For the place operation, the sphere and tool have the highest adhesion because they have been observed by the EB. The place strategy is to have a shearing trajectory that breaks the contact between the tool and sphere without breaking the contact between the sphere and surface.

Feddema and Simon [6] focus on visual servoing for the assembly of LIGA parts. Morishima [7] explores manipulation of biological particles via dielectrophoretic forces and laser tweezers. Kasaya [8] describe the manipulation and handling of micro spheres, roughly 30 mm in diameter, through a combination of electrostatic forces and complex manipulation commands. In order to increase volumetric production, Bohringer [9] explores ultrasonic vibration to eliminate surface adhesions, coupled with electrostatic forces to position and align parts in parallel.

Figure 1: Two tweezers hold a cell while two scissors perform an operation

Lasers are an attractive option for noncontact micromanipulation. Lasers have an enormous range in power and exposure time: the power can range from 10-3 to 1015 watts per square centimeter while the exposure time can range from 10-15 seconds to several hours. For cutting applications, laser scissors have high power and short exposure times while for manipulation applications, laser tweezers have low power and long exposure times. In a recent review paper, Berns [9] discusses current and future applications of optical tweezers and scissors. Scissors can produce a chemical change that inactivates a chromosome in a living cell and the alteration can persist in the cloned progeny of the cells. Scissors can cut a micron sized hole in a cell membrane (that seals within a fraction of a second) and tweezers can insert (or remove) molecules or genetic material into the cells. Berns speculates that within a decade cellular surgery would be possible in which tweezers hold cells while scissors delete a faulty gene and remove genetic material (see Fig. 1).

Since 1996, Stelzer and Hörber (and their colleagues) have been developing the photonic force microscope (PFM). The PFM can both take 3D images of living cells (at a much higher resolution than 2D light microscopes) and measure physical properties (forces and viscosity) at the molecular level. The PFM consists of a small (nm or mm) bead (latex or gold) that is trapped by an optical tweezer. The position of the bead relative to the geometric focus of the laser trap can be measured with nm spatial and ms time resolution along all three dimensions. As the bead moves along the exterior (or interior) it can become attached to the cell or the contents of the cell. Careful experiments can measure the range of the force of attraction. If the bead is coated with antibodies and binds to a particular protein, the PFM can measure the motions and forces created by the protein. Recent papers on the PFM include measurement of viscosity [11], four independent measurements of the spring constant [12], and measurement of z position using the intensity of forward-scattered light [13].

Approach

To focus our micromanipulation research, we will develop intelligent control systems for two paradigm problems: (1) micro and nano manipulation and (2) automated assembly of micro machines. Some of our recent research efforts have included lab automation and micro-array technologies for rapid functional analysis in support of the human genome program [14,15,16]. Since 1989, we have been developing modular software to control autonomous robotic systems. The robotic systems have included a large mobile manipulator, two vehicles with novel wheels, and large manipulator designed for precision motion and force control. The essence of an intelligent system is sense, decide, and act. Robotic systems require sensors, computer controlled machines, and control software. The modular control software has four components: sensory processing, world modeling, behavior generation, and value judgment. Sensory processing works with the world model to find the best current estimates of the state variables that describe the world. Behavior generation consists of three tasks: job assignment, planning, and execution. The value judgment module evaluates progress toward the goal and identifies error conditions. The control software has a hierarchical structure with planners, and controllers at each level. The high levels control the overall process while the lower levels control process details and operate at high speed. We plan to design and build the control modules a few at a time. We will perform experiments to evaluate the performance of the modules, identify sources of error and failure, perform basic research, and design the next generation of the system. We have found that the analysis and correction of unanticipated errors has been our most fruitful research strategy (To quote Isaac Asimov: "The most exciting phrase to hear in science is not "Eureka!" but rather "hmm, that's funny.").

Micro and nano manipulation. We will establish an experimental laboratory that will include all of the necessary hardware and software to perform a wide range of manipulation experiments. An inverted optical microscope will be both a viewing platform and our experimental theater. Manipulation, sensing, and sampling will be accomplished by the PFM, optical tweezers, optical scissors, and a mechanical probe all operated under piezoelectric control. For nano manipulation, we will need access to an Atomic Force Microscope (AFM).

When our laboratory is established, we will begin our experimental program. To make this proposal an independent research unit, we will describe micromanipulation experiments on living cells. When the total CESAR research program is established, we will define a series of experiments in nanomanipulation. Potential micromanipulation experiments include: mapping the surface of cells, using antibodies to locate cell membrane proteins, measuring the pH and extracting samples from inside cells, extracting chromosomes from cells, inserting materials (genetic, chemical, or physical) into cells, and measuring the restoring forces for protein structures. Our three year goal is to create an autonomous control system that will insert the PFM bead into a cell, map the inside of the cell (while avoiding getting attached to the organelles in the cell), and remove the bead from the cell.

Automated assembly of micro machines. The second task in this program is directed towards the manipulation and mechanical assembly of parts (micro components) that range in size from 50 to 1000 mm with sub-micron details and clearances. This investigation will focus on novel manipulation methods that can support automated assembly in the micro-scale environment. Our present infrastructure consists of a four degree of freedom, force reflecting micro-teleoperation system with sub micron motion resolution. We will leverage off of the experimental and theoretical experience we have in teleoperated micro-assembly to establish a foundation for automated assembly of micro-sized machines. Bohringer [9] points out that microassembly is performed by humans with tweezers and microscopes or with high precision pick-and-place robots, both of which are serial processes. However, one of the fundamental advantages of many micro fabrication processes is the ability to produce large batches of parts. Our fundamental goal is to first understand the basic mechanics of micro-assembly and then explore advanced methods for batch assembly of micro components in an endeavor to produce low cost, high throughput assembly of micro machines.

Deliverables and Milestones

Proposed major milestones for the basic research tasks described in this section are given below. All work will be performed on a best-effort basis.

FY 01: Examine cell surface and analyze cell contents. We will establish a system to examine and sample tissue culture cells growing in a micro titer plate. After exposing cells to chemicals, we will examine cell surfaces by PFM to detect changes. A mechanical probe (a fine capillary) will be moved near a cell that is being observed by the PFM, the cell will be ruptured by laser, and a sample of the cell contents will be taken into the capillary for analysis. All processes will be automated. (09/01). Insertion and planar assembly. There is presently very little quantitative information related to the forces experienced during micro assembly of parts of various sizes and shapes. We will focus on analyzing, theoretically and experimentally, the range and magnitude of forces experienced during assembly of micro components. Our emphasis will be on friction, compliance, electrostatic and adhesion forces during the different phases of part interaction and assembly. We will leverage off of our existing micromanipulation infrastructure for proof-of-principle experimentation. (09/01).

FY 02: Extract intracellular components. The goal is to locate and remove a chromosome from a cell. The object to be removed will be located by the PFM and the material will be extracted using a combination of optical scissors, optical tweezers, and the mechanical probe. (09/02). Advanced part manipulation. Assembly of complex micro-sized systems will require more than planar manipulation. There is presently no way to accurately position and orient micro sized parts for assembly. We will explore advanced forms of micromanipulation directed towards controlling both translational and rotational motion. Topics will include advanced forms of actuation (of parts serially as well as in parallel), sensing and control. (07/02).

FY 03: Measuring the restoring forces for protein structures. By varying the laser intensity, we will use both the PFM and the optical tweezers to measure forces by either interacting with or pulling on molecules. (03/03). Explore the inside of a cell with the PFM. At the present time, the particles used for imaging with PFM tend to get stuck within the cell. We will develop an intelligent control system that can map the inside of the cell without getting attached to any component of the cell. (09/03). Array based assembly. The greatest potential for high throughput production of micro and millimeter sized systems lies in parallel micro-assembly. We will explore the potential for parallel part manipulation and assembly through fixturing, arrayed actuation, or imposed electrostatic or ferrohydrodynamic fields directed towards self-assembly, such as those referenced by Bohringer [9] and Blums [17]. (09/03).

Furthermore, we will deliver to DOE/BES one copy of each publication submitted to the open literature, and one copy of each invention disclosure submitted for patent. Major technical highlights will also be provided to DOE/BES on a timely basis.

References

CESAR - Center for Engineering Science Advanced Research Oak Ridge National Laboratory

Disclaimers    -