Description of Instruments and Programming Needs

Beebe Group STM Project – September 2004

Goal

To produce a new computer program or macro that will allow the STM tip and sample to be approached to one another in a few minutes, while checking for “in-range” tunneling current signals.  The tip-sample gap will initially be a few mm, and after a successful approach operation, it will be less than one nm, and the tip will not have been crashed into the sample.

Components of the System (see associated diagram)

• PC operating in Windows 98 (new P-III motherboard with PCI and ISA slots)
• SPM32 commercial software program to control all instrument scanning and data acquisition
• Together with STM Control Unit, SPM32 software can send TTL signals to an appropriate controller in order to approach tip to sample.  The software already includes finely tuned routines to check for “in range” signals, incorporating time delays, and other features necessary to accomplish a successful tip-sample approach.
• Digital I/O card in ISA slot (Data Translation DT-2821-G-8DI).  We anticipate that the program to be developed will make use of this board to drive the Inchworm Controller.
• 8 differential A-to-D inputs with 12-bit resolution, 0 to ± 10 V, 250,000 samples per second.
• 2 D-to-A outputs with 12-bit resolution and programmable output ranges of 0 to ± 2.5 V, 0 to ± 5.0 V, 0 to ± 10 V, 130 kHz maximum throughput
• 16 channels of 8-bit digital I/O (TTL)
• External trigger with 100 – 1,000 nsec pulse width
• 4.00 MHz on-board clock
• Dedicated 3COM Ethernet card for communications with STM Control Unit
• RS232 Serial port for communication with Inchworm Controller
• STM Control Unit (RHK model SPM 100)
• Together with SPM32 software, can send TTL signals to another appropriate control unit in order to approach tip to sample. 
• STM scan head and preamplifier (home built)
• Includes the inchworm motor (Burleigh/EXFO model IW-800 series) capable of making small steps to approach the tip to the sample without crashing.
• Preamp (RHK Technology, model IVP-300) measures STM tunneling current when tip is in tunneling range (less than one nm from sample).  Preamp gain is 1 V per nA.  SPM32 program monitors and samples this tunneling current continuously.
• Inchworm Controller (Burleigh/EXFO model 8200)
• Includes some sample programs in Visual Basic to control the controller and inchworm motor.
• SPM32 program already produces a TTL signal that was used to drive older generation of this controller.  Current version of Inchworm Controller requires RS232 signals to control motion of inchworm motor.

Schematic Diagram of System Components

Generalized Steps Required for Computer Program

Type 1:  Stand-alone program

1. Program queries user as to whether an approach or retract operation is desired.
2. If an approach is desired, program initiates an approach operation.
3. Program checks STM Control Unit for tunneling current.  If no current is sensed, STM is considered out of range, and approach operation gets a green light.
4. Program sends appropriate signal to Inchworm Controller to cause tip to move closer to sample.
5. Program again checks STM Control Unit for tunneling current.  If no current is sensed, repeat step 3.  If current is sensed, go to step 5.
6. Stop approach motion of Inchworm Motor and turn over control of STM Control Unit to SPM32 program.

Type 2:  Program or Macro running within SPM32

1. Program queries user as to whether an approach or retract operation is desired.
2. If an approach is desired, program initiates an approach operation.
3. Program checks STM Control Unit for tunneling current.  If no current is sensed, STM is considered out of range, and approach operation gets a green light.
4. Program sends appropriate signal to Inchworm Controller to cause tip to move closer to sample.
5. Program again checks STM Control Unit for tunneling current.  If no current is sensed, repeat step 3.  If current is sensed, go to step 5.
6. Stop approach motion of Inchworm Motor.

Important Issues to be Considered

1. Moving the tip will always cause a small amount of fake transient currents due to voltage changes near the tip (dV/dt), and the tip’s stray capacitance (C_stray).  The induced current I_ind is proportional to C_stray × (dV/dt).  Variable time delays following the motion of the tip, and prior to the sampling of tunneling current, will have to be built in and optimized so that fake signals do not confuse the process of checking for the “in-range” condition.
2. The threshold value of the tunneling current used to decide what is “in-range” should be adjustable.
3. The Inchworm Motor works by using a combination of clamping and unclamping operations on clamps located at each end of a moving shaft (1 and 3 in the diagram).  The shaft will hold the STM tip on one end (not shown).  The clamps are connected by a piezoelectric element (2 in diagram) capable of very fine motion.  Between the clamping of one end and the unclamping of the other end, fine steps are used to move the shaft in a given direction to extend or retract the center piezoelectric element when only one clamp is clamped.  In the diagram at right, there is net motion of the shaft toward the right.  The size of these fine steps can be adjusted from ~ 0.1 nm to ~ 1,000 nm.  Current checking, as described above in (A.) must be performed after each such step.
4. At the end of shaft motion in a given clamp cycle, a clamp/unclamp operation will be required.  Current checking will also be required before and after these more drastic clamping operations.  Because of the more drastic nature of these clamping and unclamping operations, a larger disturbance in the induced current I_ind is likely to result.  Time delays must be built into the program to allow these false signals to decay to zero before checking the current.  These signal-sensing delays will generally need to be longer than for the fine steps described in (C.) above.  These “clamp delays” will need to be independently adjustable from the “step delays”.
5. Although one solution would be to take billions of 0.1-nm baby steps, thus avoiding crashes, such a solution would be very time-consuming.  Therefore, the program should be able to intelligently adjust its step size depending on how close the tip is to the sample.  This might be related to the magnitude of the induced current measured prior to reaching an in-range condition, although this is not necessarily true.
6. While the Inchworm Motor is approaching the tip to the sample, the STM’s feedback system will independently be looking for tunneling current, and it will be ready to pull back if it finds any.  Although this extra feedback can be considered as a kind of insurance policy against accidental tip crashes, it has to be remembered that the feedback system has a limited response time (on the order of a kHz), whereas the preamp and current-measuring circuits have a larger bandwidth (on the order of 6 kHz).  The speed of all inchworm motions should take into account the finite response time of the feedback loop on the STM.