Author(s):
Nebjosa Jaksic -
Colorado State University
Melinda Szabo - Colorado State University
Implementing an AFM for less than $10,000 based on a commercially available scanner, piezo driver, photodetector, the National Instruments Educational Laboratory Virtual Instrumentation Suite (NI ELVIS), and NI LabVIEW graphical programming environment.
AFMs are widely used in nanotechnology to image and manipulate a variety of metallic and nonmetallic nano-size particles. However, they are expensive with prices ranging from about $50,000 to more than $500,000. We developed a cost-effective AFM for this project consisting of a commercially available scanner, a piezo driver, a photodetector, and a PC equipped with NI ELVIS and running LabVIEW software.
An AFM scans a surface by touching it with a sharp tip on the end of a cantilever and creates an image of that surface based on the tip deflection information. With sharp tips with diameters in the range of a few nanometers and cantilevers capable of detecting forces of a few piconewtons, nanometer sized surface features can be imaged. Also, nano-size objects can be manipulated. AFMs are essential in biology, chemistry, biochemistry, and medicine research.
AFMs are used to examine and verify research results at the nanoscale level. Materials that can be investigated by AFMs include thick and thin film coatings, ceramics, composites, glasses, synthetic and biological membranes, metals, polymers, and semiconductors.
Today, commercially available AFMs command high prices, which largely limits their use. Many small schools and research institutions cannot afford them. Widening the accessibility of AFMs required a new cost-effective design.
Design Considerations
An AFM consists of a coarse approach mechanism that brings a sample within a few micrometers of a scanning tip; a precision scanner to scan the sample surface in nanometer increments; drivers to direct the cantilever and tip in the x, y, and z directions while scanning; a controller to control the x, y, and z motions; a mechanism to detect deflection data from the cantilever; and a data acquisition system to process and display the data under software control. To obtain high-resolution images of a large area in the order of tens of micrometers in just a few minutes, an AFM has to move the tip over the surface quickly and must have a fast and precise data acquisition system.
Our previous experience has shown that some data acquisition and control boards from other vendors are fast and precise enough for the project; however, we chose National Instruments products because they were more user-friendly. Since the Colorado State University-Pueblo engineering department just started its mechatronics program, the selection process was underway for computerized data acquisition and control equipment for the Mechatronics and Controls Laboratory. Since the requirements for the laboratory and for the AFM were similar, 10 NI ELVIS workstations with LabVIEW were selected for the laboratory. One of the NI ELVIS workstations was used for the AFM development.
AFM Hardware
A scanner assembly by Agilent coupled with a stepper-motor driven worm gear coarse approach mechanism and a quad-diode (four-quadrant) photodetector are housed together and placed on top of an NI ELVIS development board. This in turn is placed on a heavy metal plate held by four elastic cords attached to a steel frame for improved mechanical vibration isolation. To complete the instrument, there is an NI ELVIS PC control card and the PC. The scanner assembly consists of a 3D actuator (a piezoelectric tube that can be expanded or contracted and bent in any direction), a small laser producing a light beam of a few millimeters in diameter, and a cantilever attached to the 3D actuator with a tip on the bottom and a reflective surface on the top. The laser light reflects off of the cantilever top. The quad-diode photodetector captures the reflected light beam, thus measuring the deflection of the cantilever.
Data Acquisition and Control System
The control system is divided into three subsystems: a coarse approach, a fine approach, and a scanning subsystem. The data acquisition system obtains data from the four-quadrant photodetector, records it, and then displays the data.
The coarse approach control subsystem controls a 400 steps/revolution stepper motor through four digital outputs on the NI ELVIS board (DO0, DO1, DO2 and DO3). A LabVIEW virtual instrument (VI) is used to control the motor. Using this VI, the stepper motor via the worm gear and a precision screw assembly moves the scanner assembly towards the sample until the tip touches the sample.
The fine approach control subsystem controls the z-axis of the piezotube actuator. Since there are only two 16-bit DACs on NI ELVIS boards, they are reserved for the x and y piezotube motions. Control of the z-axis is accomplished by the 12 V programmable power supply with a maximum resolution of 7-bits.
The scanning control subsystem controlling the piezotube’s motions in the x and y directions uses the two 16-bit DACs on the NI ELVIS board to perform the scanning function. Theoretically, using a 40-micrometer scanner with a 16-bit control, one can achieve a resolution of about 0.6 nanometers per step. Now, the sharp tip diameter becomes the limiting factor of the AFM resolution. The DACs are capable of outputting 10 V. Since the piezotube requires higher voltages for operation, the voltages from the programmable power supply and the DACs are multiplied by 10 before they are fed to the piezotube.
The data acquisition subsystem consists of a four-quadrant photodetector by Advanced Photonix, Inc., four low-noise operational amplifiers TL082AC, and four channels of the ADC on the NI ELVIS development board. The four photodetector quadrants are sending voltage signals via operational amplifiers to the NI ELVIS board. From these signals the position of the laser spot on the detector can be determined, and the laser spot on the detector can be represented using an x-y chart on the front panel.
Software Implementation
To control the AFM, a flowchart is created, divided into 16 steps, and then programmed using LabVIEW to a standard state machine where each state executes the code and determines which state to transition next. VIs are created and tested for each state, then they are incorporated into the main LabVIEW program together with a variable initialization VI.
Amazingly, after about six months of development, a successful AFM was built at about one-tenth of the cost of its commercial counterparts.
Author Information:
Nebjosa Jaksic
Colorado State University
2200 Bonforte Blvd.
Pueblo, CO 81001
United States
Tel: (719) 549-2112
Fax: (719) 549-2519
n.jaksic@colostate-pueblo.edu
Print
PDF
