The microcantilever of silicon with or without an integrated
tip is now established as a tool useful for investigating
many signal domains. In the most common instance it is used
to monitor surface topography. In other areas where a tip is
not used, the cantilever is used as a transducer with
unprecedented sensitivity. It is a calorimeter with a
sensitivity measured in femtojoules. It is a magnetometer
with an improvement of several orders. It is a device for
monitoring molecular reactions and measuring the mechanical
properties of single molecules. These areas will be
emphasized in this presentation.
[I6.002] Scanning Probe Microscopy - Tools for Manufacturing & Development
H. Kumar Wickramasinghe (IBM, T.J.Watson Research Center, P.O.Box 218, Yorktown Heights, NY 10598)
Scanning Probe Microscopy has evolved as an important tool
in both science and technology. This talk will outline some
of the work done by the author and his group in developing
scanning probe techniques for manufacturing and development.
The Magnetic Force Microscope (MFM) was developed to
non-destructively image magnetic properties of heads and
disks with 20 nm resolution. The Electrostatic Force
Microscope (EFM) and its extensions - such as the Kelvin
Force Probe - allow us to image dopant profiles, charge and
workfunctions all on the nanometer scale. Force Microscopy
is now routinely utilized in semiconductor manufacturing
lines as a high precision metrology tool in the control of
critical levels. Representative examples from all these
areas will be presented.
[I6.003] Single spin magnetic resonance force microscopy: Progress and challenges
Daniel Rugar (IBM Almaden Research Center)
This abstract not available.
[I6.004] Novel Lock-In Waveform Techniques for Measurement Signal-to-Noise Ratio and Dynamic-Range Enhancement in Highly Noised Experiments
Andreas Mandelis (Photothermal and Optoelectronic Diagnostics Laboratories, Dept. of Mechanical and Inudstrial Engineering, Univ. of Toronto, Toronto M5S 3G8, Canada)
Problems with the ability of conventional single-ended lock-in amplifier (LIA) measurement techniques to detect weak variations in noised signals are well-known to experimental physicists. They are mainly due to two limiting instrumental factors: the poor signal-to-noise ratio (SNR), and the instrumentally limited signal magnitude (or amplitude) dynamic range (DR). The SNR is limited by the variable frequency-scan noise bandwidth and the necessary instrumental transfer function normalization. The DR is limited by the output signal baseline, which may be too high to detect relatively small variations introduced by the presence of weak processes or inhomogeneities. In this talk (of potential interest to all experimental physicists), I will introduce two repetitive input LIA waveform methodologies which effectively address these universal measurement problems: The pulse-duration-scanned LIA rate-window, and the lock-in common mode rejection demodulation. Neither of the two techniques requires instrumental transfer-function normalization, unlike the conventional 50% duty-cycle or sinusoidal-modulation frequency-scan techniques. The former waveform gives superior SNR for signals with very low power content, such as in high-frequency thermophysical measurements of thin films using laser infrared techniques. The latter waveform employs a more complicated periodic modulation (e.g. of a pump laser beam) to provide a differential-signal-equivalent LIA demodulation through common-mode rejection. Several case studies will be presented involving noised thermal-wave detection in solids or liquids with (otherwise immeasurably) small perturbations away from the homogeneous state. Comparisons with conventional single-ended LIA and/or pulsed techniques will demonstrate the substantial power of these novel universal measurement methodologies in achieving high-detectivity in experimental systems under adverse conditions limited by poor-SNR and/or low DR. These encompass commonly used instrumental table-top physics set-ups, where conventional frequency- or time-domain measurements are very difficult or impossible.